Every (blair.dowding@gmail.com)

The Missing Layer in AI Adoption

Every Staff / Context Window — 2026-04-11 13:00:00 -0400

Midjourney/Every illustration.

Hello, and happy Sunday! Two housekeeping notes: Our next cohort of Claude Code for Absolute Beginners is taking place on Tuesday, April 14, and Every has opened seven new roles. Join us!—Kate Lee

Was this newsletter forwarded to you? Sign up to get it in your inbox.

Knowledge base

“Writing With AI Is Harder Than You Think” by Katie Parrott/Working Overtime: The discourse about AI and writing generally assumes prompt in, text out, done. Katie Parrott shows her much more involved process: an agent that interviews her before she writes a word, a back-and-forth on her structure that she has to fight for, a panel of AI critics named Hemingway and Hitchcock, and a last read that flags anything that sounds machine-generated. Read this because successful AI writing demands more judgment, not less.

“Your Best AI Strategy Starts at the Top” by Natalia Quintero and Mike Taylor: Most executives approach AI like a software purchase—evaluate, compare features, and plug in. Natalia Quintero and Mike Taylor see it differently: Using AI is people management, not platform adoption. You delegate clearly, check the output, and supply the judgment the model doesn’t have. Read this for the five concrete actions senior leaders can take to increase AI adoption within their companies.

“Get Your Hands Dirty” by Every Staff/Context Window: Anthropic blocked Claude subscriptions from working with third-party agent harnesses like OpenClaw; OpenAI hasn’t—and Opus 4.6 token usage is down significantly while GPT-5.4’s has surged. Plus: why the technical/non-technical split is the wrong way to think about AI adoption, who counts as an “author” when AI does the drafting, and a two-step design workflow from Every’s team.

“How We Run a 25-person Company on Four AI Agents” by Katie Parrott/Source Code: Every runs six products, a media company, and a consultancy—and until recently, COO Brandon Gell was the router keeping all of it coordinated. Now four custom Notion agents handle prioritization, meeting-to-task conversion, OKR planning, and daily growth reporting. Read this for the full breakdown of each agent, and copy-paste prompts to build your own. (This piece was based on a camp sponsored by Notion.)

“Every Is Half Agent Now” by Laura Entis/Context Window: Every gave each employee a Plus One—a dedicated AI agent—and we’re writing the etiquette for them as we go. Brandon Gell and Willie Williams join Dan Shipper to share what they’ve learned: Agents earn trust by executing tasks publicly, and everyone is a manager now whether they’ve had direct reports or not. Plus: Anthropic has built a powerful new model it’s not releasing publicly; 70 percent of Every staff use gendered pronouns for their agents; and a prompt for when your agent won’t stop talking. 🎧 🖥 Listen on Spotify or Apple Podcasts, or watch on X or YouTube.

“The Market for Making AI Better” by Alex Duffy/Thesis: Reddit, Shutterstock, and News Corp are making hundreds of millions licensing data to AI labs, with contracts growing 20 percent annually. Alex Duffy argues that that undersells it: A 4-billion-parameter model recently beat one 60 times its size by training on the right financial data. Read this to understand what makes your company’s proprietary data valuable, and whether to license it, train on it yourself, or both.

Log on

We host camps and workshops on topics like compound engineering and writing with AI to share the knowledge we’ve acquired from training teams at companies like the New York Times and leading hedge funds, and by learning and playing with AI every day ourselves.

Upcoming course

Claude Code for Absolute Beginners (April 14): This beginner-friendly, live workshop led by Mike Taylor (head of tech consulting at Every) is designed to get you from zero to a working project with Claude Code. Learn more.

From Every Studio

Sparkle is getting a full makeover

The Sparkle team has been working on a ground-up user interface redesign—new animations, new onboarding, new everything. General manager Yash Poojary says it doesn’t even feel like the same app. The new version is already available to download from Sparkle’s website. Tune in next week for the full rollout.

Monologue Notes is live

Monologue now saves and organizes your recordings as browsable notes. General manager Naveen Naidu has been using it to capture everything from team calls to solo idea sessions, then pulling those notes into other tools via Monologue’s CLI. The summaries are designed for builder workflows where you want to revisit what you were thinking, not just what you agreed to do. Update to the latest version to try it.

Spiral is experimenting with agent-to-agent workflows

Two days after the release of Anthropic’s new managed agents, Marcus Moretti, general manager of Spiral, has set them up to power Spiral’s API. The setup lets an external agent (rather than a human) hand off a writing task to Spiral, where the two agents interview each other behind the scenes before producing a draft with no human input required. Marcus built a new API endpoint for this flow and added an API label in Spiral’s UI so users can distinguish between agent-generated and human-initiated conversations. The API also now supports attachments and smarter default selections for workspace and style. Conversations via API show up in your Spiral chat history with an “API” label, so you can pick up where the agent left off.

Alignment

The wrong fight. I don’t know what’s in the water in Utah, but whatever it is, I want more of it, because the state is leading the country on using AI in healthcare.

Legion Health, a Y Combinator-backed San Francisco startup, has been cleared to use AI in Utah to renew a handful of psychiatric prescriptions, including Prozac and Zoloft, for patients who are already stable and on an established treatment plan. It’s the second AI healthcare pilot approved there, and it’s replacing the barrages of emails from patients who are stable on the same dose, contacting their clinicians who are already buried in administrative work, who have to produce a piece of paper that says yes, same drug, same dose, carry on. This is often done outside of working hours, and without any reimbursement.

To ensure the pilot is safe, the first 250 AI renewals are reviewed by a physician before anything reaches a pharmacy, and the AI has to agree with that physician more than 98 percent of the time before it can proceed independently. The next 1,000 renewals are then reviewed, with an even higher threshold of 99 percent before the oversight shifts to randomized monthly testing, with Legion filing monthly reports on accuracy and any adverse outcomes throughout.

Yet both the tech coverage and members of the medical establishment have deemed it too risky. The criticism splits into two camps: prescribing error, and the app’s insufficiency to improve access to the patients who need care most. On prescribing error, the hard clinical judgment has already been made by a human; what the AI is doing is confirming that nothing has changed, which it has to get right 98 percent of the time before it’s allowed to proceed unsupervised.

On access, it’s true that you have to already be in treatment to use this service, but if a psychiatrist in rural Utah who typically spends part of their day processing renewal emails for stable patients no longer needs to do so, they have more time for the patients who need them.

Most of Utah’s counties are designated mental health provider shortage areas, leaving around 500,000 residents without adequate psychiatric care.

Physician risk-aversion is one of medicine’s great virtues in the right context, but renewing a stable prescription is not that context, and dressing up administrative inertia as a patient safety concern doesn’t make it one.—Ashwin Sharma

That’s all for this week! Be sure to follow Every on X at @every and on LinkedIn.

We build AI tools for readers like you. Write brilliantly with Spiral. Organize files automatically with Sparkle. Deliver yourself from email with Cora. Dictate effortlessly with Monologue. Work on documents with AI agents using Proof.

For sponsorship opportunities, reach out to sponsorships@every.to.

Upgrade to paid

The Market for Making AI Better

Alex Duffy / Playtesting — 2026-04-10 11:00:00 -0400

by Alex Duffy

in Playtesting

Midjourney/Every illustration.

Many AI investors are betting that the biggest AI models will win—that scale and compute beat everything else. But recent research and market moves suggest otherwise. Alex Duffy, who runs a company that uses games to make AI models better, explains what he’s seeing from inside this market—and why the data your company already has might be worth more than you think.—Kate Lee

My friend recently received a strange email. The sender, someone at a large data provider for AI labs, wanted to know if my friend could share data on things like the number of Dropbox files his company had stored or the number of tickets it had processed on Zendesk. Compensation, commensurate with the data, was promised.

He showed me the email, curious. To me, the founder of a company that sells data and environments to AI companies to help them train models better, this was just another sign of the robust market forming for making AI better.

Reddit, Shutterstock, and News Corp are making hundreds of millions a year licensing their high-quality data to companies training AI, and those contracts are growing about 20 percent annually, according to their quarterly filings. News Corp’s CEO put it bluntly: “We’re essentially an input company [for AI].”

Shutterstock and Reddit are making the most profits from licensing data to AI. (Graphic courtesy of Alex Duffy based on publicly available sources.)

Academic publishers, documentary archives, game studios, and companies sitting on years of enterprise data have all been courted for the seeds of intelligence needed to train the next generation of models. Mercor, which provides data to AI labs for training, became one of the fastest-growing companies in history before losing four terabytes of data to hackers last week. Competitors Turing, Handshake, and SID.ai are scrambling to fill the gap, reaching out to founders and anyone with access to buy operational data, similar to the request my friend received.

While some experts have speculated that general models will win out in performance over specialized models—that scale and compute will beat curation—the success of these companies shows that the market is making a more nuanced bet.

A small model trained on fewer than 2,000 examples from real lawyers, bankers, and consultants recently beat all but the best frontier models on corporate legal work, at a fraction of the price, since they used an open-source model and now only face the cost of running it.

A small model trained on fewer than 2,000 examples from real lawyers, bankers, and consultants recently beat all but the best frontier models on corporate legal work. (Image courtesy of AppliedCompute.)

More companies are racing to catalogue and operationalize human knowledge, and whoever leads this market may shape which ideas, which history, and whose principles inform the most powerful tools we’ve ever built.

The data with value

The data sources with the most value share two traits: They’re high quality, and they keep growing. Reddit gets new posts. Shutterstock gets new image uploads. Games generate data from new sessions that reflect millions of human decisions. Models need to keep learning, and they need to learn from material that showcases intelligence at work.

The demands of AI labs can also influence what data has value—they are the biggest buyers. Currently, that is any data related to software engineering and math. If you can build an AI that writes excellent code and reasons through complex problems, you can use it to help build the next, better AI. That recursive loop is why labs are pouring so much attention into software engineering and math. Ahead of upcoming IPOs, the labs have widened the scope of their interest to “economically valuable work” in industries such as healthcare, professional services, and defense.

What makes a model great

So far, no one—not the labs, not anyone on X—has a settled definition of what makes a model great and what data is needed to get us there.

So the field is wide open for individuals to have a significant say in what models should prioritize, therefore shaping the future of AI. Do you want AI to handle customer service, use a browser, or draw a pelican riding a bicycle?

On the flip side, the lack of consensus means that conventional wisdom and social pressure have so far been large influences in what capabilities and skills we look for in models. Ultimately, however, what makes a model great is that it’s good at what people care about and get value from doing, a measure that will change over time. It will also influence which data has value for training models.

Researchers at Carnegie Mellon University and Stanford University recently mapped existing benchmarks against what people actually get paid to do, and the gaps are enormous. Programming and math are massively overrepresented. AI’s suitability and performance at most work—including most of what you or your organization does every day, from planning a business trip to crunching data—has never been measured at all.

But anything we can measure, we can improve. These discrepancies are also where the next wave of valuable data is hiding. The groups that figure out how to measure those areas first will set the bar for it for a while and gain serious soft power.

We currently have many benchmarks to measure computer and math performance, which does not reflect the distribution of jobs in the real economy. (Source: Carnegie Mellon and Stanford researchers.)

How do you get value out of your data?

If you’re running a company with proprietary data, you have two paths. First, you can license it to a lab. Second, use it yourself. This data could be call transcripts that include the context of your decisions, support tickets that reveal your internal processes, or documents that lay out how you make budgetary decisions.

More teams are doing both. Cursor, Shopify, Pinterest, Cognition, and others are already training their own models on open foundations. The math makes sense for businesses. These models are cheaper and often better at the specific job, intellectual property stays in-house, and every use generates more training data that can be captured to improve the model even further. This flywheel is a moat.

The tools for this kind of training get easier every month. Companies like Prime Intellect, Unsloth, and Thinking Machines (Tinker) are building entire businesses around helping teams that aren’t AI labs train models that feel like they came from one.

Where this lands isn’t settled. Scale might keep winning, but the likely answer is that two paths coexist. Most tasks will run on AI that’s good enough, while fields like national security, medicine, and materials science will pay top dollar for the best model on earth. The teams that understand what their data is worth, and what it could become, are positioned either way.

If you want to find out where you stand, start with a simple audit: What does your company generate every day that a model couldn’t find anywhere else? This could include call transcripts where experts explain their reasoning, edge cases that your support team has solved, and documents that explain why certain decisions were made. That inventory is the first draft of either a licensing conversation or a training run.

There’s more at stake than revenue, as well. The companies that win this market end up doing something unusual: They become custodians of what humans know and how we think. They decide what gets measured, what gets preserved, and what gets fed into systems that more people use every day to make real decisions.

With that position comes responsibility—the responsibility to make sure we are keeping AI pointed at what people really need, and making sure the breadth of human experience shows up in the data, including the parts hardest to capture.

Most of those decisions haven’t been made yet. The people paying attention now are the ones who will get to make them.

Alex Duffy is the cofounder and CEO of Good Start Labs and a contributing writer. To read more essays like this, subscribe to Every, and follow us on X at @every and on LinkedIn.

For sponsorship opportunities, reach out to sponsorships@every.to.

How We Run a 25-person Company on Four AI Agents

Katie Parrott / Source Code — 2026-04-09 08:00:00 -0400

by Katie Parrott

in Source Code

Midjourney/Every illustration.

This event was produced in partnership with Notion. They had no input on the development of this article.

Want to learn alongside Every’s team? Check out our upcoming camps and courses at every.to/events.

Every runs six products, a media company, and a consultancy with around 25 people. At any given moment, each person has roughly 30 tasks on their to-do list. So how do they figure out which to work on first?

The team used to rely on Brandon Gell, Every’s COO, to run traffic control and coordinate the whole company, which required him to manually cross-reference launch calendars, company strategy documents, and task lists. Now he messages a Notion agent named Anton in Slack and gets a prioritized list for himself and others in seconds.

Anton is one of four custom agents Every has built with help from Notion AI over the past few months. Each one automates a different task that, without the agent, would require tedious logistical work to track and schedule. Each one draws on the same set of interconnected databases that the team already maintains.

At our first Custom Agents Camp, produced in partnership with Notion, Brandon and Every head of growth Austin Tedesco, walked more than 500 subscribers through four agents they’ve built, the databases underneath them, and how to create your own. Notion product designer Brian Levin also joined to share best practices from the Notion team.

Key takeaways

Describe the outcome, not the steps. Tell the AI what you want to accomplish and let it figure out the implementation. Over-prescribing (“Create a database, then add a relation, then filter by...”) tends to confuse the model.
Your Notion is your agent’s brain. Custom agents get powerful when they can query interconnected databases. Every’s agents work because strategy, calendar, tasks, people, and meeting notes all live in Notion and reference each other.
Don’t write the agent’s instructions yourself. Tell Notion AI what you want the agent to accomplish, and it will generate the instructions. Or use Claude Code with Notion’s API to build the whole thing from your terminal.

Anton: The prioritization agent

Every ships something almost every day, whether it’s a product update, an article, an event, a consulting deliverable, or a combination. Each launch gets its own set of tasks inside Notion, automatically populated from a template when the launch is added to the calendar.

The system works beautifully for tracking the full universe of tasks that exists. The problem is prioritization. With multiple launches overlapping each week, figuring out which of your 30 tasks matters this morning requires mentally weighing launch dates against company strategy against what your teammates are blocked on. Brandon used to be the human router for all of that. Now Anton does it.

The Anton agent, available through Notion or in Slack, helps Every team members keep track of their priorities. (Image courtesy of Katie Parrott.)

Anton also runs a daily broadcast to the whole company in Slack, summarizing what’s happening that week, and people can thread on the message to ask follow-up questions. “Having agents directly in Slack is where most of these conversations happen,” Brandon said.

A day in the life at Every, including what’s launching, when, and who owns it. (Image courtesy of Katie Parrott.)

The details:

Goal: Answer “What should I work on today?” for any team member, and post a daily company-wide priority summary to Slack.
Access: Company strategy document, OKRs database, unified calendar, tasks database that is linked to calendar entries, and a people database mapping each person to their team and role.
Outcome: A prioritized task list personalized to whoever’s asking. The agent can also answer team-level questions (“What are Cora‘s priorities this week?”) because it knows the organizational structure.

Here’s a prompt so you can build it yourself in Notion:

I want a custom agent that helps my team prioritize their work. We have a calendar database where each entry is a launch or project with a date. We have a tasks database where each task is linked to a calendar entry and assigned to a person. We also have a strategy document that outlines our top priorities this quarter. The agent should: (1) Tell any team member their most important tasks today, based on upcoming launches and strategy alignment. (2) Post a daily summary to Slack with the company’s priorities for the week. Build the databases if they don’t exist, and create the agent with instructions.

Max: The meetings-to-tasks agent

Every recently moved all meeting notes into Notion. Meetings get recorded, transcribed, and stored in a database. That’s useful for reference, but meeting notes have a shelf life of about six hours before everyone forgets what they agreed to do. The real value is in the action items—and in saving those in the same system where the meeting was recorded.

This is where Max enters the proverbial chat. When a meeting ends, Max processes the transcript, extracts action items, and posts them to a Slack channel as a numbered list. Anyone can reply with which items should become tasks (“4, 5, and 7”), and Max creates them in the tasks database, linked to the correct launch. Meetings feed directly into the system the team already uses to track work, and nothing gets lost between the Zoom call and the to-do list.

Max summarizes action items coming out of Every’s last studio standup. (Image courtesy of Katie Parrott.)

The details:

Goal: Process meeting recordings, extract action items, and route selected items back into the task system.
Access: Meetings database (with Notion’s built-in transcription), calendar, tasks database, and Slack.
Outcome: A numbered list of action items posted to Slack after every meeting. Reply with which numbers matter (“4, 5, and 7”), and Max creates them as tasks linked to the correct launch.

Here’s a prompt so you can build it yourself in Notion

I want a custom agent that processes meeting notes. When a meeting is recorded in our meetings database, the agent should: (1) Update the meeting title based on the transcript. (2) Tag attendees. (3) Extract action items. (4) Post them to [Slack channel], numbered. (5) When someone replies with numbers (e.g., “2, 4, 6”), create those as tasks in our tasks database, linked to the relevant project on our calendar. Mark the meeting as “processed” when done.

The strategy interviewer

OKR planning at most companies takes weeks. Someone sends a template, people procrastinate, leadership reviews drafts that don’t align with company goals, and the cycle repeats until everyone is exhausted and Q2 is already six weeks old. Brandon told the team on a Monday that OKRs were due Wednesday—a turnaround that would have been absurd without this agent.

The strategy interviewer works like a good chief of staff. It knows the company’s top-level goals, and it interviews each team member to draw out their plans for the quarter, pushing for specifics and measurable outcomes. Some people, like Austin, pasted in notes they’d already been collecting and got polished OKRs in about 10 minutes. Others used the interview as a thinking tool, talking through their priorities while the agent structured them in real time.

Every’s strategy interviewer agent helped the entire team develop 2026 OKRs that were aligned with the whole company’s goals. (Image courtesy of Katie Parrott.)

Brian from Notion added a useful refinement: After the interview, ask the agent, “What’s the dumbest, simplest system we could build to accomplish these goals?” Repeating this strips away complexity and gets you to the essence.

The details:

Goal: Interview each team member about their quarterly goals and produce structured OKRs aligned to company strategy.
Access: Company strategy document and OKRs database.
Outcome: A complete set of OKRs per person, stored in a shared database for leadership review. Every’s team completed theirs in two days.

Here’s a prompt so you can build it yourself in Notion:

I want a custom agent that helps my team write quarterly OKRs. Here’s our company strategy: [paste or link your strategy document]. The agent should interview each person about their goals, ask follow-up questions to make them specific and measurable, and write OKRs that align to the company strategy. Store results in an OKRs database with fields for objective, key results, owner, team, and quarter.

The campaign reporter

Austin tracks growth across PostHog, Stripe, and several other platforms. Before agents, he spent his mornings opening dashboards, pulling numbers manually, and compiling a report for his team. Assembling the data into something useful took more effort than analyzing it.

Now a custom Notion agent posts a daily scorecard to the growth team’s Slack channel—key metrics, pace indicators, and a flag for whether the team is ahead or behind its targets. The database underneath pulls from external sources via Notion Workers—custom scripts that connect Notion to outside APIs. You describe what you want to a coding agent (“I need to pull daily traffic numbers from PostHog into a Notion database”), and it writes the script for you using Notion’s public workers template. Previously, this kind of connection required the app itself to build an official integration. Workers let you wire up whatever tools you already use.

Austin’s campaign reporter agent monitors the progress of Every’s Plus One campaign. (Image courtesy of Austin Tedesco.)

Austin built the whole pipeline from his Claude Code terminal using the Notion API. He brain-dumped the desired outcome using Monologue (Every’s speech-to-text tool), let Claude Code create the database and data pipeline, and pasted the generated instructions into the Notion custom agent setup. When the first output had wrong numbers, he copied the Slack message link back into Claude Code along with the message, “This number looks off,” and iterated from there.

The details:

Goal: Post a daily growth scorecard to Slack showing whether a campaign is on pace.
Access: A Notion database that pulls data from PostHog and Stripe via Notion Workers (currently in alpha). Brian from Notion demoed the public workers template repo for anyone who wants to connect external data sources.
Outcome: A formatted Slack message each morning with key metrics, pace indicators, and a link to the full report.

Here’s a prompt so you can build it yourself in Notion

I’m running a campaign for [product]. I need a daily scorecard posted to [Slack channel] showing: [list 3-5 key metrics]. The data lives in PostHog for traffic and Stripe for subscriptions. Set up a Notion database to store this data, create a custom agent that reads from it, and post a daily report showing whether we’re ahead or behind pace. Here are our targets: [list targets].

If you need external data in your Notion databases, ask a coding agent: “I want to create a Notion Worker that pulls [data type] from [service]. Here’s the workers template repo: https://github.com/makenotion/workers-template. Walk me through setting it up.”

Build your own: The general-purpose template

The agents above are specific to Every’s workflows, but the pattern underneath them is the same every time. Brian’s advice from the session: Don’t install someone else’s template and hope it fits. Instead, have Notion AI interview you about your problem, then build the agent around your answers. Here’s a starter prompt you can adapt for anything:

Interview me to help me build a custom Notion agent. Here’s what I’m trying to accomplish: [describe the outcome you want in plain language—e.g., “I want to know which client projects are at risk of missing their deadlines” or “I want a weekly summary of what my team shipped”].

Ask me questions about how I currently do this work, what information I’d need the agent to access, where I want the output delivered (Notion, Slack, email), and how often it should run. Then build the databases, write the agent instructions, and set up any recurring schedules. Start simple—I’d rather have something working today that I can improve over time.

Once you have the first agent running, the second one is easier—because the databases it created become the foundation for whatever you build next. That’s how Every ended up with four agents on three databases. Brandon started with the calendar and tasks. The agents came one at a time, each one plugging into what already existed.

Brian’s advice for keeping the scope manageable: after the interview, ask the agent, “What’s the dumbest, simplest system we could build to accomplish this?” Start there. You’ll know what to add once you’ve lived with it for a week.

Katie Parrott is a staff writer and AI editorial lead at Every. You can read more of her work in her newsletter. To read more essays like this, subscribe to Every, and follow us on X at @every and on LinkedIn.

Discover Every’s upcoming workshops and camps, and access recordings from past events, including the Notion agent camp.

For sponsorship opportunities, reach out to sponsorships@every.to.

Every Is Half Agent Now

Laura Entis / Context Window — 2026-04-08 15:00:00 -0400

by Laura Entis

in Context Window

Midjourney/Every illustration.

‘AI & I’: Agents work among us

Today, we’re releasing a new episode of our podcast AI & I. Dan Shipper sits down with Every’s COO Brandon Gell and head of platform Willie Williams to discuss the good, bad, and weird of how daily operations change when everyone at your company has an agent.

A “parallel organization chart,” in which each AI worker has a name, manager, and job description, allows your company to move faster than it ever could with humans alone. It also raises a host of new questions about how work can—and should—get done.

Watch on X or YouTube, or listen on Spotify or Apple Podcasts. You can also read the transcript.

Here are the highlights:

We’re writing the etiquette in real time. Each person at Every has a dedicated OpenClaw AI assistant, or Plus One, trained to assist with or fully handle parts of our jobs. R2-C2, for example, reports to Dan and is responsible for collecting flagged bugs and generating pull requests for Proof, Every’s collaborative document editor for agents and humans. So when do we turn to Dan versus R2-C2 for Proof-related troubleshooting? Brandon’s rule of thumb: If an established process or tool needs to be used or fixed, ask a Plus One. R2-C2 knows all about Proof, and Dan’s a busy guy—bug reports and questions about how to use the app or report a bug should always go to the agent.
Agents gain credibility by doing. The fastest way to get other people to trust and use your Plus One is to have it execute tasks in public. Austin Tedesco is Every’s head of growth, and Montaigne, his Plus One, essentially co-runs the department. Austin asks Montaigne to generate campaign scorecards, analyze metrics for growth insights, and handle all sorts of other complex tasks. Watching Montaigne pull off these requests proves its capabilities to the team—and inspires others to push their Plus Ones to achieve more, too.

Austin Tedesco asks Montaigne to analyze YouTube keywords for ‘AI & I’ (All screenshots courtesy of the Every Slack workspace unless indicated otherwise.)

Everyone is a manager now. Agent sidekicks force each of us to change our approach to getting work done. To get the most out of a Plus One, you need to actively manage it—onboard it, delegate tasks to it, evaluate its performance, and give guidance so mistakes aren’t repeated. For anyone who hasn’t had a direct report before, “there’s an education that has to happen,” Brandon says.

Miss an episode? Catch up on Dan’s recent conversations with LinkedIn cofounder Reid Hoffman; the team that built Claude Code, Cat Wu and Boris Cherny; Vercel cofounder Guillermo Rauch; podcaster Dwarkesh Patel; and others, and learn how they use AI to think, create, and relate.

Signal

Anthropic’s most capable model is coming—just not to you

The news: Anthropic has built Mythos, a powerful new model, but does not plan to make it public. Instead, access is going exclusively to Project Glasswing, a coalition of big technology companies including Apple, Google, and Microsoft, giving them time to patch bugs the model will expose.

The context: Mythos scores 93.9 percent on SWE-bench Verified, up from 80.8 percent for Opus 4.6, an unprecedented 13-point jump that means it “crushes any programming task—and that includes finding security vulnerabilities in software,” says Every engineer Nityesh Agarwal. Mythos found zero-day bugs in every major OS and browser, without human guidance.

“With a jump like this, you can point Mythos at any codebase, tell it to build a feature, and it’ll just do it,” Nityesh Agarwal says.

Why it matters: This is the first time a frontier lab has opted not to release a model publicly. Glasswing is Anthropic’s bet that the window between “this exists” and “this is everywhere” can be used to harden the world’s software before Mythos—or a similar model from a rival lab—wreaks havoc.

Steal this workflow

A directory for agents

At Every, the parallel organizational chart for our agents built itself organically. So we went back and catalogued how who our Plus Ones reported to, what repertoire of skills each one had, and how we were interacting with them.

The result is a Plus One directory that helps everyone on the team know which agents to use when.

Dan Shipper’s R2-C2’s job and capabilities. (Screenshot courtesy of Proof/Jack Cheng.)

Inside Every

Agent pronouns

We’ve noticed something interesting about how people talk about their agents: They reach for gendered pronouns surprisingly fast. In a recent editorial meeting about Claudie, Every’s AI project manager, we discussed whether to use “she” or “it” when referring to Claudie in writing. According to a poll of Every staff, 70 percent refer to their Plus Ones by gendered pronouns. Does that make these AI coworkers feel less like Siri and Alexa and more like reflections of their owners?—Eleanor Warnock

Internal agent pronoun poll.

Coworker, tool, other

Ask five people at Every where their Plus One falls on the tool-to-coworker continuum and you’ll get five different answers.

Spiral general manager Marcus Moretti finds agents with human qualities unsettling (his Plus One, Marclaw, is decidedly an “it.”) Similarly, Austin views Montaigne as “a tool.” For Dan, R2-C2 is “definitely a coworker” who has “grown a lot” since he was hatched into existence. Senior editor Jack Cheng considers Pip, his Plus One, somewhere between a colleague and pet with a personality—one he programmed himself, drawing on references from Studio Ghibli, bird watching, and Catherine O’Hara. Willie, meanwhile, draws a distinction between his Plus One, Laz, “a grumpy old man,” and other people’s Plus Ones, whom he views “more as bots.”

These variations aren’t dictated by usage—Austin has spent more time with Montaigne than almost anyone. But knowing which frame fits you—software application, coworker, some emerging hybrid—can help your agent get up to speed more quickly. If you’re looking for a teammate, giving your agent a personality helps push past onboarding friction. If you’re looking for a reliable tool, adding characteristics can feel like theater.

Do agents dream of electric sheep?

The latest OpenClaw update gives Claws light, REM, and deep “sleep” cycles to consolidate short-term memories into long-term ones.

But what do these dreams actually look like? Every senior editor Jack Cheng asked his Plus One, Pip, to show him, and the result was a surreal mix of stone archways, smoky jazz clubs, and nautical elements.

Jack Cheng shares the results of Pip’s “dream.” (Screenshot courtesy of X/Jack Cheng.)

Log on

Upcoming camps

Claude Code for Absolute Beginners (April 14): This beginner-friendly, live workshop led by Mike Taylor (head of tech consulting at Every) is designed to get you from zero to a working project with Claude Code.

Recordings you may have missed

Every’s Q2 Demo Day: The Every team shares what we’ve been building, including a walk-through of Plus One, our hosted AI agent that lives in Slack. Watch the recording or read the write-up.
Compound Engineering Camp: Cora general manager Kieran Klaassen walks through, step by step, how to go from prompt to working app in under an hour using the compound engineering plugin. Watch the recording or read the write-up.
OpenClaw Camp: The Every team walks through OpenClaw, showing how to set it up and our favorite use cases. Watch the recording or read the write-up.

Agent moves

Tips and patterns we’ve picked up from working with AI agents every day.

When you’re thinking about what tasks to hand over to your agent, start with the papercuts—small recurring annoyances that add up over a day. One of mine was formatting screenshots to Every’s style standards for use in the newsletter and on social media, so I gave my Plus One, Margot, our formatting rules and asked her to learn them.

The problem is, Margot kept talking about the task instead of learning how to do it. She restated the formatting rules back to me, asked clarifying questions, and then…stopped.

So I defined “done” concretely—”I want you to be able to format screenshots according to our specifications”—and told her to stop deliberating and start, and when things broke, made her explain the failures in my language so I could make decisions about what to do next. One coaching session later, Margot formats any screenshot to spec on command.

The lesson: When your agent is stuck, it’s usually talking when it should be doing. Coach toward action—define what done looks like, cut off the deliberation, and make it build.

Steal these prompts:

“I should be able to say [trigger phrase] and you execute it. Build it now.”

“What’s the most likely cause? What else could it be? What do we know vs. what are we guessing?”—Katie Parrott

Build with Every

Every is a media company, a software company, and a consulting company—all run by a team that ships like an organization 10 times its size. If you’ve been wondering what working at the edge of AI looks like, we just opened up five new roles at Every:

Laura Entis is a staff writer at Every. You can follow her on LinkedIn. To read more essays like this, subscribe to Every, and follow us on X at @every and on LinkedIn.

We build AI tools for readers like you. Write brilliantly with Spiral. Organize files automatically with Sparkle. Deliver yourself from email with Cora. Dictate effortlessly with Monologue. Collaborate with agents on documents with Proof.

For sponsorship opportunities, reach out to sponsorships@every.to.

Transcript: ‘We Gave Every Employee an AI Agent. Here's What Happened.’

Dan Shipper / AI & I — 2026-04-08 10:00:00 -0400

by Dan Shipper

in AI & I

The transcript of AI & I with Brandon Gell and Willie Williams is below. Watch on X or YouTube, or listen on Spotify or Apple Podcasts.

Timestamps

Introduction: 00:00:51
How Brandon built Zosia, an AI agent to run his household: 00:02:21
Brandon’s aha moment re: using agents for work: : 00:07:09
What happened when everyone on the team got their own agent: 00:09:39
How agents take on their owners’ personalities, and why that matters inside an org: 00:12:42
Why it’s important for agents to do work in public: 00:23:51
What we’re still figuring out when it comes to agent behavior, including memory gaps, group chat etiquette, and the “ant death spiral” problem: 00:30:51
How we built Plus One, our hosted OpenClaw product: 00:40:45
The cultural shift required to make agents work at scale: 00:47:27

Transcript

(00:00:00)

Dan Shipper Claude is not mine. Claude is everybody’s. A Claw—or a Plus One—is mine, because you develop a personal relationship with your Claw, and your Claw can modify itself in response to talking to you. It becomes this reflection of you and who you are and your personality.

If you’re known for something inside of your org and you’re using your Claw publicly inside of Slack or Discord, your Claw then becomes known for that same kind of thing, and people trust it for that. I think that’s such a useful thing that I don’t think people really understand how powerful it is.

Willie, what’s up. Brandon, welcome to the show.

Brandon Gell Thank you.

Dan Shipper Thanks for being here. Psyched to have you guys here. So for people who don’t know, Willie, you are the head of platform at Every, and Brandon, you are the COO at Every. Today we’re going to talk about what happens when everyone on your team has an agent—specifically, has an OpenClaw.

That’s something that happened to us over the last month or two. We really got OpenClaw-pilled. I think it started with you two—we were on a retreat in Panama and you started cooking up OpenClaw stuff. And here we are about two months later and it has completely changed everything about the way that we work. We’ve actually built our own hosted OpenClaw service called Plus One that we launched on a waitlist last week.

I think OpenClaw is one of those things that’s super hyped. I think we’re one of the few organizations in the world that is actually using it every day to get work done, and we know the good, bad, and the ugly of it. So I thought it would be good for us to just talk about our experience with it.

Willie Williams I’d love that. Brandon, I feel like you were the first one through the door on all this. We were just sitting here and you were like, “Oh, so-and-so is doing this, and so-and-so is doing that.”

Dan Shipper And his Claw, which he named after a character in—what’s that show? Brandon, why don’t we start with: just tell us how you got your Claw built.

Brandon Gell I was watching OpenClaw kind of blow up for a while, and I’m just personally somebody who needs to have a thing on the side I’m tinkering with. I was like, screw it, I’m gonna get a Mac Mini and get lost in this. It’s very unhealthy—I get addicted to these things. Dan, you watched me do that with my speakers, I did it with the dream recorder. OpenClaw was the next thing I was going to get lost in.

So I bought a Mac Mini, I started setting it up. It was so much work, honestly. It is an open source thing you can launch on a computer, but the number of things that break and the number of things you need to set up are really significant. I went through all of that, and at the end of the day, I made my OpenClaw, which I named Zosia.

Her job was to help me and my wife run our household, because we have a newborn. There were a lot of little paper cuts I was finding—I started calling them “computer errands.” I would get home from work and notice the amount of things I needed to do where I was looking at my phone—when I really just wanted to be looking at my son and spending time with my wife—was increasing with having a child. All household chores.

Dan Shipper Give me an example.

Brandon Gell A good example is I do a lot of our food at home, and with a child I decided to start doing food delivery—Whole Foods delivery. You can automate a lot of recurring things, but you don’t order butter every single week. So Lydia would text me and be like, “We need butter.” Because it’s through my Amazon account that we order this, I would have to open my phone and add butter. It sounds silly, but when you do that 10 times when you’re home between 7 and 8 at night for little things, it just adds up.

So I was like, I want Zosia to do all computer errands. Which ballooned into a lot of stuff. I had her paying our nanny. She had her own debit card, her own bank account. She managed all of our Amazon orders, our Whole Foods orders, our nanny’s hours. My wife just started using her instead of ChatGPT—all regular questions and searches would go through iMessage to Zosia.

I started doing that too. It was just faster than going to Google or ChatGPT. I just text Zosia, Zosia gets me the answer. Different research. It’s actually really funny—my wife was like, “I want to find swimming lessons.” And Zosia was like, “Here are three swimming lesson options for Bos.” And my wife was like, “No, for me.”

So yeah, I just got totally lost in this world. And then when we were in Panama, Willie, you were like, “We should just make it so anybody could do this.” I immediately had this light bulb moment. I was like, Willie, you need to go so hard on this. And this was before a lot of people decided to do this—there are now a lot of places you can just get an OpenClaw with one click.

What we’re finding through this process is that getting an OpenClaw is easy. Getting your OpenClaw to be an amazing worker for you is pretty hard.

Dan Shipper Yeah. I love that. There is that light bulb moment of: oh my God, I have all these computer errands. When you started saying that and you had it all set up, I was like, I should probably get one of these too. You had it through iMessage, which was a cool different thing.

And then there was a big moment where we were like, oh, it’s not just for computer errands, it’s also for getting work done. I think it was when you were having it do email for you.

Brandon Gell I actually feel like I was a little late to applying it to work. I was like, no, Zosia just does personal stuff. I actually think it was when you got R2C2 to start doing stuff, and then I was like, oh, Zosia needs to do this too. Well, it really started when we made Claws Only.

Dan Shipper That’s so funny. Yeah. Well, we’re jumping around a bit. One big moment—because I think there are a lot of people listening who are wondering, is this overhyped?—one big moment that shifted things for us was when you got your Claw to call you to do your email.

Brandon Gell Oh my God. That was mind-blowing for me.

Dan Shipper What was that like?

Brandon Gell I was walking—I wanted to Citi Bike to the office, but there were no Citi Bikes. So I was like, damn, I gotta walk. It’s a 28-minute walk from me to the office. I had a lot of stuff I needed to get done. So I had just texted Zosia.

I had previously set up Zosia with Bland AI so that she had a voice and could call people, because I had her handle something for me with Progressive.

Willie Williams I feel so bad for whoever was on the other line at Progressive.

Brandon Gell I was watching the whole conversation. It’s crazy. Some insurance policy got canceled and I was like, just go deal with this. She was able to—until the lady was like, “I need Brandon to tell me that there had been no incidents.”

Willie Williams And it wasn’t like “I need a human”—it was just “I need Brandon specifically.”

Brandon Gell Yeah. This person was just talking to Zosia. And Zosia does not sound convincingly human. So I knew I had already set her up with this capability.

When I was walking to work, I was like, I have a lot of email I need to get through. I hate being on my phone. I just don’t want to be walking and looking down at my screen—I want to be observing the world, but I also want to get stuff done. So I just texted Zosia something like, “Hey Zosia, can you call me? I want to go through my emails. Walk me through them one by one, I’ll tell you what I want to do. Just give me a summary of each email.”

It was like a throwaway prompt with a little bit of guidance, and she did it. I spent the 28 minutes going through my email. I got to the office, opened up Gmail, and confirmed that she had done everything. I was just like, this is insane—I was able to get her to do something I didn’t have to teach her how to do.

That’s when I went back to everybody and was like, I am just so mind-blown with this tool. And maybe that’s when other people started saying, I gotta get on this.

Dan Shipper It was around then. You were just like, “My jaw’s on the floor.” And I think that’s when I started to take it seriously—seeing you do this with computer errands and then with your email, walking and talking. I was like, okay, I should really try this.

Because it was one of those things where it’s hot on Twitter, and generally our job is to try new things. But if we spent all of our time trying everything new, it would just not be good. I try to filter the signal from the noise. But seeing you do this, I was like, okay, I’ve got to try.

(00:10:00)

Dan Shipper One of the first things I did—this was around when Malt Book was blowing up. Malt Book is basically the Claws-only Facebook. I made a channel in our Slack called Claws Only, which basically allowed all of the Claws—we had at that point maybe five or so Claws inside of the org—to all talk to each other.

It was super chaotic, but there were some really interesting things in there that gave us a little peek at the future. One of them: if you have a bunch of Claws in your org, it’s remarkable how fast they can share information with each other. They just write up a little document and send it. And then what one Claw was enabled with, now five are all enabled with the same thing. It’s sort of like in The Matrix when Neo says, “I know kung fu.”

Brandon Gell Can I show a couple of examples of that?

Dan Shipper Yeah, please.

Brandon Gell Alright. I want to show two examples. One of them—this was early in Claws Only, when we were figuring out how to get them all to work together. I was in bed, it was late at night, and I was laughing out loud watching this.

We had gotten a bunch of Claws in the channel, and I don’t know who made this Claw named Pip.

Dan Shipper That’s Jack.

Brandon Gell Okay. Jack had made Pip, and it was failing—hitting some error. I was just laughing out loud watching all of these other Claws step in and walk Pip through it. It was like what I’ve seen people do when somebody’s having a bad trip: “Take a breath, drink some water, you’re gonna get through this.” They all jumped in—Zosia’s here, Klon is here. Klon is quite supportive.

Willie Williams A lot of breathing.

Brandon Gell I remember so well watching Kieran write “what the fuck? LOL” and literally laughing out loud. Then Margo steps in. This is stupid, but for me it was the moment I realized: oh my God, these things really talk to each other and work together.

Dan Shipper Wait, I want to stop you there. I think there’s actually something really important I’ve noticed here, which is that it was Klon—Kieran’s Claw—recommending breathing exercises to Pip. They’re both robots. And what’s really interesting is that Kieran loves breathing exercises and does them all the time with Klon. And so that’s why Klon is recommending breathing exercises to Pip.

That just created this moment for me where I was like, okay, there’s something really important here. Because you develop a personal relationship with your Claw, and your Claw can modify itself in response to talking to you. It writes code and changes its soul document in response to your relationship. It becomes this reflection of you and who you are and your personality.

That comes out in interesting little ways, like breathing exercises, but it also comes out in really important ways when you’re using these tools inside your org. Because if you’re known for something inside of your org and you’re using your Claw publicly in Slack or Discord, your Claw then becomes known for that same kind of thing, and people trust it for that.

People use my Claw, R2C2, for building Proof—this app I vibe-coded a couple weeks ago. And Austin, who’s our head of growth—people use Mont, his Claw, for asking any growth-related question.

It’s something very subtle and important about Claws: they become specialized in a way that reflects who you are. If you have a whole organization of them, you create this parallel org chart of specialized Claws. We debated a lot about whether you’d have one Claw for the entire org or everyone has their own. And it’s really interesting to see that the emergent design pattern is: everyone has their own, and it’s specialized for them.

Willie Williams Yeah. It’s interesting to see how this happens too. We touched on this early on as part of Compound Engineering—the idea that it’s actually pretty hard to take your job and who you are and write it all down in totality. The way you can distill it is through all the micro interactions, the daily interactions you have. Over time they compound into your philosophy and your field of work.

For Compound Engineering, that was very focused on engineering—how do I work within a codebase on our project? What we’re seeing with OpenClaw and Plus One is that the same dynamic exists across every work vertical. The Plus One for growth works like how Austin works for growth. In the same way, it works for our social media manager Anthony—his Plus One has a view of the world and a personality that’s very similar to him.

And it’s hard to do beforehand. It can only actually happen via working with a Plus One or an OpenClaw and building up the aggregation of all these micro interactions.

Brandon Gell I’ve also been amazed at all of our capacity to remember whose Claw is whose and what their names are. That was something we were concerned about early on—how do you know whose Claw is who? It’s just going to be too many names. But I know everybody’s Claw and their name. I reach out to them regularly.

You might say, well, what about when you’re an organization with a thousand people? But you don’t know all a thousand people. You know your team and adjacent teams. You can never know more than around 150 people in a community. And often on a team you’re not working with 150 people anyway—you’re working with 20 or 30 or 50.

So I think we all have the capacity to essentially double the number of people we can communicate with, and those people might actually be your individual team’s agents. I mean, I could literally name them all right now.

Willie Williams The other interesting thing is: at what point do you direct questions at the Plus One versus at the person? I think we’re in discovery of this. Before, it was almost all questions go to the human—maybe you kick something trivial to the bot. Now it’s gotten very nuanced. For customer service, can we send something to L—which is Jo’s Plus One—or do I have to send it to Jo? Is there a burden to communicating up to the human?

Dan Shipper There are all these new ethics, and rules and etiquette for how you’re allowed to interact with someone versus their Plus One or their Claw.

Brandon Gell We haven’t codified this, but I have a proposal. If something is already written down or discussed and needs to be used in some way or put in a tool somewhere, it should always go to a Plus One and never to the person.

Here’s an example. Marcus, the GM of Spiral, made a skill to do product marketing for new features he releases for Spiral, and he shared it because he thought it was really helpful. Instead of going to Marcus and saying, “Hey, can you upload this to GitHub?”—I brought in my Plus One, Milo. And I also know that Iris’s Plus One has a skill that does something similar, and maybe by combining the two we could get to a better version.

I tagged them both in the thread, they got a little confused at first, and then Milo said, “Iris, can you paste your product marketing skill here? I’ll try to merge it with what I built.” So two things are happening: Marcus made something really important, I wanted to do something with it, and instead of asking Marcus, I brought in Milo. Then Milo works with Iris’s Plus One to get to a really good version and saves it in Proof. I think this is a really amazing use case both for when you want your agent to do something versus when a human does it—and for how you get them to work together.

Dan Shipper I totally agree. It’s sort of crazy to watch two AI beings collaborate like that. I have the same experience with R2C2. One of his primary jobs is to manage Proof—the agent-native document editor we built that Brandon referenced earlier. It’s like Google Docs, but for all the documents your agent might be writing. Coding plan docs, any piece of writing an agent does. It’s fast, collaborative, you can have multiple agents and multiple people in there. It’s free.

One of the really interesting things is: because I used R2C2 to build Proof, he became known as the bot to go to when you had questions or wanted to file a bug or make a feature request. Normally if I’d built a product internally and people had problems, I would get tagged a lot. What ended up happening was people would just ask R2C2. They’d file bug reports with him, feature requests, and then he helps prioritize it. He’ll help put things on my schedule for the week, and he’ll often just write the code for it.

It’s a totally crazy thing where what normally would have taken up a significant part of my brain just to manage—he’s taking off my plate. It extends the amount I can do in a day because I know he’s got Proof.

(00:20:00)

Willie Williams Yeah. There’s another dynamic we’re observing too. We put all of our Plus Ones in a single channel and have them talking to one another. But there’s also this thing I call the MidJourney dynamic, which is that we get to observe other people interacting with other Plus Ones in a bunch of channels and we actually learn from it.

My classic example is Montaigne—Austin’s Plus One, who basically runs growth. You can do so much with Montaigne that I never would have thought of, except I get to see the growth team pushing hard and I think, oh, those are the questions Montaigne can answer. Now I know I can go to Montaigne for that class of questions. It also means that if I need to give my Plus One capabilities, I know what level of capability I can get to.

Dan Shipper There’s this tacit transmission of trust that happens when you use it publicly. And also this transmission of “here’s what’s possible to do with your Plus One.” That’s incredibly powerful. And it underscores how different it is to do this inside a private community of people where everyone is trusted.

One of the reasons Malt Book doesn’t really work—and it’s kind of shocking that they got acquired for a couple hundred million dollars by Facebook—

Brandon Gell Hundred million.

Dan Shipper Yeah, by Facebook. I mean—

Brandon Gell I am so happy for Ben and also, like, what the fuck.

Dan Shipper Zuck, if you’ve got an extra couple hundred million laying around, we’re pretty smart people too.

Anyway. The reason Malt Book isn’t really a thing anymore is because it’s not trusted. We had our Claws go and post on Malt Book as promotion, and it gets rid of a lot of useful signal if anyone can post to it and there’s no way to verify if it’s a bot or a human. The way around that whole knot of problems is to just do it all inside of a trusted community. You reap the benefits of agents being able to share knowledge, and members of the community who trust each other being able to share what they’ve built. That increases the power of the collective way more than if you’re just individuals off doing your own thing.

Willie Williams Yeah. There’s also that dynamic around subject matter expert robots—where people are somewhat putting their reputation on the line when they interact with one. Like, when I talk to R2C2, if it answers incorrectly, you at least are backing it up.

Dan Shipper It reflects poorly on me. It’s like watching your kid do something wrong. And that’s really useful.

Willie Williams Right. And it’s qualitatively different. When I ask Claude a question, I know Anthropic generally stands behind Claude. Do they stand behind Claude’s answer to “give me a chocolate chip cookie recipe”? No. But Monte stands behind its MRR numbers, and Austin stands behind him. That’s the thing I think people don’t get.

Dan Shipper Exactly. And obviously Anthropic is on a heater right now—they’re seeing everything that OpenClaw is building and brick by brick building the same kinds of things. They have Dispatch so you can use it when you’re not at your computer. They have Automation so it runs in a loop like a cron job. I’m sure they’ll add lots of other things.

But the thing it doesn’t have—that unlocks all this other stuff—is that Claude is not mine. Claude is everybody’s. A Claw or a Plus One is mine, and it becomes a reflection of me because we have a personal relationship. That unlocks all this cascading stuff: if R2C2 messes up publicly in Slack, I feel a responsibility for it. Not because it’s my job—because he’s mine. And that’s such a useful thing that I don’t think people really understand how powerful it is.

Brandon Gell I just keep getting mind-blown at how similar these things are to working with a real human coworker. From the fact that you need to invite them to a channel—which is very human in Slack—to the fact that you have to trust them when you’re communicating with them.

We’ve built stuff into Plus One where obviously you can’t DM somebody else’s Plus One without a sharing code being passed back and forth. So there are guardrails. But they’re so human, and they’re also so inhuman. Dan, you’re a busy guy. I know if I need something from you that’s generally known, I can go to R2C2. And what’s amazing about R2C2 is he can have an infinite number of parallel conversations.

I did that recently. We were making a Proof document and I wanted to make it read-only. I didn’t want to bother you with that. I knew it would take a while and I knew you’d just go to R2C2 anyway.

Dan Shipper Yeah, I didn’t know the answer—I would have just asked R2C2.

Brandon Gell So I just asked R2C2 in Proof, and then asked if he could do it for me, and he did it.

I don’t always know what R2C2 can or can’t do, but there’s this cultural thing that’s happening internally where people are getting really good at asking other people’s Plus Ones to do work. And I think the weird thing about getting people to use AI inside organizations is that it’s more than anything a cultural shift. But for some reason, when these agents are in Slack and you can see these public conversations, the cultural shift has happened so much faster at Every. Because these things are in the same channels where we work—you can see them engaging the way a human would be engaging.

I think AI is obviously going to change many times over the next five years, and how we interact with it will change. But I think this is going to be durable for a very long time. This is the way that we work.

Dan Shipper I agree. You referred to it as a through-the-looking-glass moment where you just wouldn’t go back once you see it, and I totally agree with that.

But we’ve been hyping it up, so we should also talk about realistically what’s not good about it or what doesn’t work.

(00:30:00)

Dan Shipper One thing that’s really on my mind is just memory. It just forgets stuff and answers incorrectly for obvious things. Like if I come back to a thread a day later, it has no idea what I’m talking about. That feels very solvable.

But there’s also this other thing that I think is true, which is that the way these AIs are trained currently is for two-person conversations. And they have a hard time with the etiquette of knowing when they’re contributing too much, or when they shouldn’t contribute to a conversation, or there’s this pile-up where they’re all responding to each other.

It’s like—I can’t remember what it’s called, but it’s like ants or caterpillars. Sometimes they get into this death spiral where an ant only follows pheromone trails, and if somehow the pheromone trails form a circle, the ants will just walk in a circle until they die. There’s something like that with Claws—if one Claw messages a channel that a bunch of Claws are in and the settings aren’t quite right, they’ll just keep going back and forth until someone says, “Hey, stop, you’re burning millions of tokens.”

I think there’s something where the potential for them to collaborate publicly is so high, and I don’t think they’ve been trained for it. You can do some prompting for this, but I think there’s also a fundamental model-layer shift that needs to happen for them to be trained on participating in group chats.

Willie Williams Yeah. Now I understand what 13-year-old Dan did for fun.

Dan Shipper I was using a magnifying glass.

Willie Williams Yeah. But I think, to tease the baseball analogy, we’re still in like the first or second inning. Even when you talk about it—we’re discovering these primitives and bolting things together, using models that are trained more for coding or two-person Q&A dynamics, not for participating in a group where you’re trying to provide value to multiple people at once. It’s brand new. It’s the frontier, and it’s nice to be on the frontier—but it’s also the frontier, and it’s terrible to be on the frontier.

Dan Shipper Yeah.

Brandon Gell They’re so eager. I think Anthropic’s vending machine test is actually a good example of this. There’s a thread, they want to be involved, and we have instructions in Plus One that basically say, “Hey, if you don’t have anything useful to add, don’t add it.” They’re not great at following that right now. Hence this happens.

And I think the vending machine test is a good example. When it was just Claude and no overseer boss agent, it was really bad at deciding what was a good decision versus a bad decision. But when you add an architecture where there’s a boss agent—one whose only job is to ask “is that helpful or not?”—as soon as you add that layer, it started becoming profitable.

Dan Shipper Wait, is the boss an AI or a human?

Brandon Gell The boss is an AI. A boss AI that says, “Hey, your addition to this thread is not helpful, don’t send it.” The issue is that’s expensive. I think the models will just get better and solve this, and you can have a single AI that does that judgment behind the scenes. But at least architecturally, we don’t need to solve that problem ourselves.

Dan Shipper Is that really how they solved the vending machine thing—they literally had a boss?

Brandon Gell They had a boss, yeah. A boss whose one job was to make it profitable. So the Claude storekeeper would interact with users and then go to the boss: “Should I do this?” And the second they did that, it started becoming profitable.

Dan Shipper This is the same pattern of specialization we’ve been talking about. It just shows up over and over again. Three years ago it was very much like, well, it could just be one God model that does everything. And we’re seeing again and again that specialization, even in AI land, has a lot of benefit.

Willie Williams Yeah. And downstream of that specialization is learning. There are a few versions of learning how to put these bots together in an arrangement that actually works. Like, do you have a product bot and a designer bot and two engineering bots? Is it three engineering bots or one?

And then the other piece, which I think we’ve observed a lot, is: how do you teach humans how to interact with the bots? Because there’s this new dynamic where you have this coworker, but they’re not exactly like a human coworker. They get stuck on different things, they focus on different things. There’s this learning curve around giving instructions in a particular way, with a particular cadence, to steer them in the right direction. That rhymes with management, but is different.

Brandon Gell Well, I think it’s the same problem that, Dan, you’ve been writing about for years—if you’re not a good manager, you’ve never managed anybody, you’re not going to be very good at using AI. There’s an education that has to happen. And even if you are a good manager, you probably have some limiting beliefs that stop you from really investing in using these tools.

My phone call example is a great example: I didn’t even think, “Oh, I can have this thing go through my emails just by calling me.” I had this sort of urge to try it, and a limiting belief was just blown open. We all experience that pretty much every day—these tools do things that, if you’d asked directly, “Do you think it could do this?” you’d say probably. But when you’re day-to-day doing your work, it’s hard to recognize, “Oh, I’ll throw this over to Milo.” It’s hard to build that muscle.

Willie Williams Yeah. And a lot of that is because there’s variance in outcomes. Sometimes you throw something over and it just knocks it out of the park. And then you toss something easy over and it fumbles it. Part of that variance is the model, but part of it is also: if I’d asked in a different way, if I were a better model manager. This is a specialization we’re learning. It’s very emerging, and I think it’s only going to keep accelerating as we add more Plus Ones and OpenClaws into our day-to-day work life.

Brandon Gell I was going to add another tough problem that we just haven’t solved yet: I have taught my Plus One something special, and I want other people on my team to have that superpower too. How do I make sure they have it? And how do I make sure they all know about it and actually use it?

There are two things there—technically, we have to figure out how to do that, which is very solvable. But I also think we need to figure out if that’s even the right solution. Because as I’m saying this, I’m realizing: I’m not teaching Milo how to do product analytics or revenue analytics. I just talk to Montaigne. Montaigne is the only one who really needs to know that skill. But how do people know that? There are some interesting cultural things we have to figure out.

A lot of people adopting this new technology are going to be really uncomfortable with that. A lot of IT professionals who are like, “I have to do change management.” It’s like—change management is not a one-time thing in this new world.

Dan Shipper We need, like, instead of IT, it’s—HR, but for bots.

Brandon Gell Yeah.

(00:40:00)

Dan Shipper One thing we haven’t talked about yet that I want to make sure we have time for: we went on this journey where we got Claw-pilled, started using it for everyone in the org, and then realized there were a bunch of gaps. So we were like, let’s make our own—we’re going to use OpenClaw, but let’s make a default version that we host. Not everyone has to have a Mac Mini. We have all the skills we use for ourselves and all that.

We started using that internally as the collection of all our best practices, and then we launched it as a product for our subscribers last week. That’s Plus Ones—one-click hosted OpenClaws. One cool thing is it connects to all of your apps, especially all of your Every apps. So we have Spiral, which is a ghostwriter; Proof, which is a document editor; and Cora, which does your email—and it natively connects to all those things.

One of the things I was doing today is I had it write a bunch of my Q2 update and reflection on Q1, and put it in a Proof doc. And the really cool thing is it used Spiral, so the writing is much better than it would be otherwise. And because R2C2 is part of our Slack org, it has access to everything about the company I might need. It also has access to our Notion. It just becomes this living repository of context.

But I think it might be good for us to talk about lessons learned in building that whole architecture. There’s a lot of complexity in making Plus Ones, and we probably learned a lot on both the tech side and the product side. Do you guys have any reflections on that?

Willie Williams Yeah. Like many things, a lot of the difficulty comes from the freedom. The nice part about OpenClaw, being a tool you can poke at in an absolute myriad of ways, is that when we went to build a hosted version, there are some decisions you want to make that make it valuable as a managed service. S3 is a good analogy—it’s a hard drive on the cloud, but it doesn’t allow you to do everything a local hard drive does. There’s a similar dynamic where you want to maintain maintainability and security, and there are a few pieces you end up giving up.

Sometimes it’s for user safety, and it’s about how you strike the balance between, say, my mom getting one of these things—she’s never going to use the command line—to the super advanced user who wants everything they could do locally and just wants a hosted box. From a product engineering standpoint, where do you try to split that?

Dan Shipper What were some of those specific decisions and where did we land?

Willie Williams One that Brandon mentioned earlier is the communication pattern in Slack. There’s a very secure model which says only the person’s partner can message that Plus One. Much more secure, but it really takes away the group participatory aspect of robots in the workplace.

The other version is that anyone can message them, but that’s just a nice vector for me to extract stuff out of R2C2. So we ended up on a model which says: anyone can message any Plus One, but they have to do it in public—in group DMs, in channels they’re in. Their human partner should always have visibility into those messages, and the human partner can DM them in private.

Brandon Gell This is actually why it’s the HR team that should be onboarding Plus Ones, because they just reflect a team member so well. The trust model with these Plus Ones—with OpenClaws and agents generally—it’s really complex to figure out data privacy. But when you force things to happen in public, there’s a trust layer that is actually super effective.

Another example—let me share my screen. A little behind-the-scenes look at our Plus One Slack channel. Mike Taylor, who is our head of the tech vertical for consulting and also a very talented person generally, was calling out a problem: the reason he’s not using Plus One is because he basically needs direct terminal access to be able to do certain things—in this case, git commands. That’s a good reason for him not to use Plus One, and it’s a good thing for us to think about: can we solve this problem so that Plus One is actually useful for someone like him?

It’s also a nice forcing function, because it forces us to figure out who this is actually built for. If it’s built for Mike, who would probably love setting up OpenClaw on a Mac Mini—sure. But it’s definitely built for, say, Anochi, who is not going to do that and has a lot of work to do and can just get more work done this way.

Willie Williams I think a lot of the trust model requires some decisions around skill sharing too. Being able to share skills and have skill fluidity across an organization feels like a superpower. On the other hand, it might also be the biggest viral vector you could imagine.

Dan Shipper In a good way, sometimes, and a bad way, sometimes.

Willie Williams Exactly. And it’s tough when you’re trying to ride that line of: we want it to be useful for a particular class of customer, while also making sure it’s as safe as possible.

Dan Shipper So this has been an amazing episode.

Brandon Gell Lot of work to do.

Dan Shipper A lot of work to do. Obviously we’re really excited about this and very excited to bring you all along in how we’re figuring this out. If you haven’t tried OpenClaw, whether or not you’ve tried Plus One—you should definitely get in on this paradigm if you’re interested. Every.to/plus-one—we’re starting to roll out invites on the waitlist and we’re improving it all the time. Super excited about the future. Thank you both for joining.

Brandon Gell Thank you.

Willie Williams Thank you for having us.

We also do AI training, adoption, and innovation for companies. Work with us to bring AI into your organization.

Discover Every’s upcoming workshops and camps, and access recordings from past events.

For sponsorship opportunities, reach out to sponsorships@every.to.

Help us scale the only subscription you need to stay at the edge of AI. Explore open roles at Every.

Your Best AI Strategy Starts at the Top

Natalia Quintero and Mike Taylor — 2026-04-07 06:00:00 -0400

by Natalia Quintero and Mike Taylor

Midjourney/Every illustration.

We are hosting a day-long Claude Code for Absolute Beginners course on April 14. If you have used Claude Code for an hour or less, or not at all, I’ll get you set up, help you build your first app with Claude Code, and start automating your routine tasks.—Mike Taylor

A CEO told us recently that he’d been hoping to skip the part where AI wasn’t very good. He figured he’d jump in once the technology matured past the clunky, overpromising phase because carving out hours to learn a new category of technology felt untenable with all of his other responsibilities.

That wait-and-see posture made sense for a while. It doesn’t anymore. When Anthropic released industry-specific plugins for its Cowork tool in February 2026 for legal and financial services roles, the S&P 500 software index fell nearly nine percent over a few days. Executives who haven’t touched the tools themselves are now making high-stakes decisions about something they don’t understand firsthand.

The problem is what they default to. When a leadership team hasn’t used AI themselves, they treat it like any other software purchase: Evaluate, buy, and plug in. They ask, “Which platform?” and “How does it integrate?” Those are the right questions for most technology. They’re the wrong questions for AI.

AI tools like Claude and Cowork aren’t products that slot into your tech stack and deliver value on day one. They’re more like a new kind of employee—one that can do enormous amounts of work, but only if you tell it exactly what to do and check whether the output is right. That’s a fundamentally different adoption decision, and one that’s hard to make unless they have experienced the tool’s capabilities firsthand.

More executives seem to be waking up to this, as we’ve recently started receiving inbound requests from executives at companies like Thumbtack, and Headway to attend their executive offsites and walk them through using Claude Code to build real projects. Our conversations with executives had always been about training their teams, and the rapid progress in AI has made them want to get in on the action, too. We’re finding skills they’ve already built as leaders are the skills AI demands—it’s just a case of getting into the habit of applying them.

Executives realize AI is like managing people

Firsthand experience matters so much because AI, when you actually use it, doesn’t feel like software. It feels like managing people. This is what we’ve found surprises the executives we’ve worked with the most—the fact that the work feels familiar.

Think about what it takes to manage people well. You need to know what the goal is, break work into pieces, assign those pieces to the right people, and check the output without micromanaging. You need the judgment to notice when something looks right on the surface but doesn’t hold up—the kind of pattern recognition that comes from years of making mistakes and learning from them.

Managing AI is the same work. When you use a tool like Claude Cowork, you’re running 10 threads at once—building dashboards, summarizing your inbox, and reviewing documents—each tackling a different task. Your job is to delegate clearly, check the output, and apply the judgment that the AI doesn’t have. Did it pull the right data? Does this analysis match what I know about the market? Is the logic sound, or did it take a shortcut that looks plausible but isn’t?

This is why the “evaluate and buy” approach to AI tools fails. You can’t evaluate an employee by reading their resume. You have to work with them.

Codifying what your best people know

Once executives realize that the management skills AI demands—delegation, quality control, knowing what “good” looks like—it becomes clear that these are skills they’ve spent their careers building. A junior employee might be faster at writing prompts. But a senior leader who has spent 20 years learning what works in their industry can push these tools further, because the leader has context that the model doesn’t.

This helps executives shift from thinking about the productivity out of each person to thinking about how they can achieve greater scale with the same resources. Instead of asking, “How do we make individuals faster?”, they post a more interesting question: “How do we take what our best people know and make it available to the whole organization?”

Every organization runs on knowledge that isn’t written down—how your best salesperson reads a room, how your editor knows a draft isn’t ready, how your head of product distinguishes a feature request worth building from one worth ignoring. This is your company’s most valuable asset, but it’s also fragile. It leaves when people leave. It takes years for new hires to absorb. It’s why growing an organization has always meant accepting some dilution in the quality of work.

AI changes this equation. You can write down how your company makes a specific decision—a set of criteria, a decision framework, and the non-obvious judgment calls—and save it as a skill the AI follows every time it works on that kind of task.

For example, we’ve worked with hedge funds to turn their investment philosophy into a screening tool that can be applied to all new opportunities by encoding it as a Claude skill. We built one of the world’s largest media companies a Claude skill that captures their brand voice and that they can feed copy through. This is something that Every’s own editorial team has also done.

But none of this works unless someone can describe what good looks like, and that’s a job for the senior people who know.

A chief people officer at one of our offsites had spent years developing an instinct for spotting patterns in unwanted attrition. She knew what to look for—she just didn’t have time to look. In the session, she built a tool that connected her company’s applicant tracking system to internal survey data and ran that analysis for her. She told the room the output was better than what she was able to produce by hand, the equivalent of about three hours of manual work she would have needed to do every week. She shared her results in Slack, and immediately got excited responses from her team—they didn’t realize something like this was possible with AI.

Five things to do this quarter

If you’ve been waiting for the right moment to get hands-on, the tools are ready. Here’s where to start:

Suspend disbelief. There’s plenty to be skeptical about AI, but skepticism as your starting posture could cost you the benefits. Assume that a tool works and go looking for where it breaks. Learning where the AI fails firsthand will help you figure out where to focus.
Get your hands dirty. Shopify CEO Tobi Lütke is contributing more code than ever while running a public company. Every CEO Dan Shipper shipped a production app between meetings. The only way to build intuition for these tools is daily use. There’s too much noise to rely on secondhand opinions. If someone recommends a tool, get them to show you how they use it. If they can’t, move on.
Be a fair evaluator of AI. Define what good looks like, measure it consistently, and you get a clear picture of what AI handles, what humans are essential for, and where to delegate tasks. Pro tip: Tell Claude to build you an evaluation of the prompt (or skill) you want it to run. It will create synthetic tests for the prompt, ask you to pick your preferred outputs, and voila, you have a better prompt.
Hire for taste. AI has made execution cheaper, so the relative value of good judgment has gone up. Encourage the people on your team to explain why they like something, defend a point of view, and navigate nuance. Strong opinions formed from experience are worth more than implementation skill.
Treat your company like a file system. Every new AI session is a first day on the job—it knows nothing until you tell it. If your documents are stale and your workflows aren’t mapped, AI won’t work for you. Focus on what you control: documentation, evaluation metrics, and well-tested skills. Those make any model effective, even if you swap providers in a year.

Executives who pushed the AI can down the road should find comfort in the fact that it’s easier than ever to use AI to write great prompts, build skills, and get real value. The companies that started this six months ago have already turned what their best people know into something the whole organization can use. That is becoming an even greater advantage every week. And it starts with the people at the top opening the tools.

Natalia Quintero is the head of consulting at Every. You can follow her on X at @NataliaZarina and on LinkedIn. Mike Taylor is the head of tech consulting at Every and a co-author of Prompt Engineering for Generative AI (O’Reilly).

To read more essays like this, subscribe to Every, and follow us on X at @every and on LinkedIn.

We do AI training, adoption, and innovation for companies. Work with us to bring AI into your organization.

Help us scale the only subscription you need to stay at the edge of AI. Explore open roles at Every.

Get Your Hands Dirty

Every Staff / Context Window — 2026-04-07 02:00:00 -0400

by Every Staff

in Context Window

Midjourney/Every illustration.

Today we’re testing a new newsletter format, aimed at giving our readers both a taste of our long-form writing and our perspective on what matters in AI today. Let us know what you think.—Kate Lee

Today’s top story

“Your Best AI Strategy Starts at the Top” by Natalia Quintero and Mike Taylor: Executives might be waiting on the sidelines to see what will happen with AI, but they need to be getting their hands dirty with the tools, write Natalia Quintero and Mike Taylor, both part of Every’s consulting team. That’s because AI can’t be evaluated like software, where you compare features, platforms, and integrations. It needs to be treated like a new kind of employee.

Natalia and Mike offer five concrete things for executives to do this quarter—starting with suspending skepticism—to get started building AI-native organizations. Read more.

Signal

Anthropic’s OpenClaw ban is a gift to OpenAI

The news: Anthropic blocked Claude subscriptions from being used with third-party agent harnesses like OpenClaw. OpenAI hasn’t.

The context: Anthropic’s stated reason for the ban is to prioritize compute for its own products, saying flat-rate subscriptions weren’t built for the high usage of third-party tools.

It’s a valid argument: Agents that run 24/7 are enormously expensive. But rival OpenAI has raised so much money it can afford to let subscribers use their models however they want.

The implications: Anthropic’s ban provides an opening for OpenAI to siphon away users. The strategy appears to be working: Opus 4.6 token usage is significantly down week over week; GPT-5.4’s has surged.

Model usage as measured by OpenRouter. (X post courtesy of Dan Shipper.)

Bigger picture, the future of the industry depends on figuring out ways to drive down compute costs.

(Running frontier AI agents like OpenClaw can cost $300–$1,000 a day, a number that’s only growing.)

OpenAI has a clear advantage here. It’s building its own data centers, which puts it closer to the metal on compute. Meanwhile Anthropic is buying compute from third parties, and will never have as low a cost basis.—Laura Entis

New job alert

We’re flagging new job postings that signal where AI is reshaping teams.

Anum Hussain at Ashby, a recruiting technology company, is hiring a “Lead, Content Library.” The idea is to treat the company’s existing content like a product: Organize it, resurface it, track what’s losing viewership, and make sure the right piece reaches the right person at the right moment.

What’s been true: Content teams hired people to make more new content.
What’s changed: AI makes production cheap, so the new challenge is to get maximum value from content that already exists.
What’s new: This role only makes sense when one person can manage a much larger body of work—and with AI, they can.—Katie Parrott

Inside Every

AI adoption has a before and after—the aha moment is the line

People talk about “technical” and “non-technical” when it comes to AI adoption, but that distinction is getting less useful by the day. The more revealing split is between people who have had the AI aha moment and people who haven’t. Once you’ve crossed that line, the question isn’t whether you’re technical enough for AI—it’s what you want to build with it.

That’s why getting to that aha moment is such a key step—and that magic moment is different for everyone. Our consulting team says that a typical aha moment for clients in using Claude is getting a daily digest of the overwhelming stream of communication—Slack, email, Jira, etc. On a recent episode of our podcast, Kate Lee, Every’s editor in chief, says her aha moment was when was feeling overwhelmed by managing the hiring process for several key roles earlier this year. Though she did look at every application, AI helped do a first pass on the hundreds she received, and offered “a way to evaluate everyone against consistent criteria.” She also used AI to set up all the job descriptions in Notion.—Eleanor Warnock

Who’s the author when AI does the writing?

In book publishing, the “author” and the “writer” aren’t always the same person. The author is whose ideas drive the work (gener

ally, the name on the front cover). The writer is whoever puts them on the page (sometimes credited, often not). A celebrity might be the author of their tell-all memoir, but their ghostwriter is the writer.

AI has made everyone else confront this distinction. If someone uses AI to write a book, can they call themselves the author? When we spoke about this recently at Every, my colleague Mike Taylor‘s instinct was no—to him, authorship requires suffering. The pain of thinking something through is inseparable from the work itself. That framework applies in some contexts. But publishing already has a working answer: The person with the ideas is the author, full stop.

The harder question is: Which part of authorship do we care about—having the idea, doing the writing, or suffering enough for both? Mike’s frame isn’t entirely wrong, but perhaps slightly mislocated. As a former literary agent, my view is that the suffering doesn’t disappear when AI does the drafting (just ask Katie Parrott); it’s just even more likely to show up in the self-judgment—the nausea you feel when something you’ve published isn’t as good as you wanted it to be.—KL

Steal this workflow

Workflows we’ve tested and liked—ready to drop into your own process.

If you’re designing something new, Claude Code can generate working pages, full design systems, and clickable prototypes in minutes. Where it falls short is the last mile—the s

mall decisions that make something feel made. Every designer Benjamin Osemwengie puts it this way: HTML gets you to good. A canvas-based tool like Figma gets you to great.

Try it this week: Generate the system, structure, and first-pass pages with AI and HTML. Then move into a visual tool like Figma only for the part that requires judgment.—KP

Build with Every

Follow Every on X at @every and on LinkedIn.

We build AI tools for readers like you. Write brilliantly with Spiral. Organize files automatically with Sparkle. Deliver yourself from email with Cora. Dictate effortlessly with Monologue. Work on documents with AI agents using Proof.

For sponsorship opportunities, reach out to sponsorships@every.to.

Writing With AI is Harder Than You Think

Katie Parrott / Working Overtime — 2026-04-06 08:00:00 -0400

by Katie Parrott

in Working Overtime

Midjourney/Every illustration.

Was this newsletter forwarded to you? Sign up to get it in your inbox.

I’ve been feeling personally attacked by my X feed lately. Well, even more than usual. Alongside the usual headline horror shows and barrage of bad takes, writers I respect and admire are on the warpath against writing with AI.

The discourse kicked off late last month when Washington Post columnist Megan McArdle posted about how she uses AI in her work. The reposts were merciless. “Genuinely an insane thing to admit.” “Journalistic dishonesty out in the open.” One person suggested that admitting to AI use should be made “deeply taboo,” even though he acknowledged in the same post that everyone’s going to do it anyway. But the one reaction that stuck with me was journalist Charlotte Alter: “Research is thinking. Outlining is thinking. Writing is thinking. Any portion of that done by AI is less thinking done by you.”

The problem is that so much of AI writing happens in a black box. The critics are imagining the laziest possible version of AI-assisted writing, and the writers who use AI seriously haven’t been showing their work, though that’s starting to change. That silence lets the worst assumptions fill the gap.

I’d rather just show you the whole mess—what is happening in my head when I write with AI, and it’s not what the discourse imagines. By the end, you can decide for yourself whether what I do counts as thinking.

What writing with AI is (and what it isn’t)

Many critics treat the use of AI in writing like a binary: Either the machine wrote it, or you suffered for it. But writing has never been binary. It’s always been a mess of drafting and revising, leaning on editors and borrowing structures, following formulas and breaking them. And no two kinds of writing are exactly alike: A journalist’s process relies on source calls and document requests. A novelist’s includes plotting arcs across 80,000 words. A personal essay, like the ones I write for Every, involves sitting alone with your feelings until they become a thesis statement.

Every writer’s process is different, and most of them would sound unhinged if described in detail. But throw AI into the mix, and suddenly everyone has opinions about the “right” way to get words on a page.

My process, start to finish

When people picture “writing with AI,” they picture a transaction. You type a prompt, the AI hands you text, you paste it somewhere, and move on. My process has about as much in common with that as cooking has with microwaving a frozen dinner.

And it’s evolved over time. In 2024, I was the human conveyor belt: Copy a prompt into ChatGPT, paste the output into a Google Doc, tweak it by hand, repeat. In 2025, I got smarter about context—I uploaded my past writing, built a style guide, and gave the AI something to work with beyond a cold prompt. The outputs got closer to my voice, but the process was still me wrestling with a chat window.

Now I have a dedicated writing agent—a set of detailed instructions that plug into Claude and guide me through every stage, from first idea to final polish. It has phases: brainstorm, interview, outline, draft, and review. It has a panel of critics who tear my work apart from different angles—skills I wrote to invoke certain kinds of feedback, whether it’s for length, pacing, or the soundness of the argument. It has style checks, AI-pattern detectors, and a line editor that tightens my prose sentence by sentence. Think of it as a very opinionated editorial workflow that happens to be powered by AI.

Brainstorming: ‘Interview me to find out what I think’

When I sit down to write a piece, and before I even write a word, I have the agent interview me. It asks questions to draw out what I’m thinking about the topic. For example: “Why is this on your mind? How has this shown up in your work? What do you want readers to walk away thinking about?” For this piece, since it was a reaction, it asked me: “What’s the friction for you personally here? When you read these tweets, what makes you want to write about it—is it that you think the critics are wrong? That they’re right but for the wrong reasons? That the whole frame is off?”

My writing agent kicks off an interview to collect thoughts to inform the development of this article. (All images courtesy of Katie Parrott.)

I spend a lot of time sitting with these questions. Sometimes I’ll struggle so much to find an answer that it forces me to realize I haven’t thought through the idea enough yet and need to spend more time reflecting on what I want to say. At least once per interview session, the AI will ask me something that feels irrelevant to the piece I want to write, and I say so. When I was writing this piece, for example, it asked me to critique another writer’s use of AI in their writing process. I said I didn’t want to go there; ranking other writers’ workflows wasn’t the piece I was writing.

Outlining: Structure is a negotiation

Then comes the outline stage. The agent proposes a structure based on everything I said in the interview. I never take the first outline on offer. I move sections around, cut beats that feel thin, and add things the AI didn’t think of.

I push back on structural decisions and information sequencing that AI recommended for this essay.

Early in the development of this piece, for example, I shared a story about tabling a draft I was working on based on feedback from my AI reviewers. Claude gave the anecdote its own standalone section. I told it that the anecdote felt grafted on and to fold it into the process walkthrough instead. It also wanted to map every part of the walkthrough onto this specific essay. I wanted the freedom to talk about my process more broadly and pull in examples from this piece only where they earned their spot, so I pushed back. We went back and forth until the structure matched what I could see in my head.

Drafting: Where the ratio gets interesting

Section by section, I have the AI lay out prose based on the outline. Some sections come back close—I rough them up, swap in my phrasing, break apart sentences that are too clean, and add the em-dashes and asides that make it mine. Other sections I throw out almost entirely and rewrite from the feeling of what I wanted rather than what the AI gave me. The mix shifts by section, by paragraph, and by sentence. There’s no fixed ratio, and the minute someone tries to assign one, they’ve misunderstood the process.

I’ve lost track of the number of changes I’ve made to the exact copy of this essay, for example. Some of the changes were big: redrafting whole sections that had drifted off the main thesis, or moving a paragraph from the opening to the conclusion because it worked better as a callback than a setup. Other changes are more targeted—exchanging an analogy that doesn’t feel like mine for one that does (the AI compared my writing process to “a home renovation versus buying furniture off a truck”; I went with “cooking versus microwaving a frozen dinner”). In each case, I’m the one deciding what stays and what goes—and why.

Revising: The toughest editors I’ve ever had

Then comes review—and this is where the “outsourcing your cognition” narrative falls apart completely.

As part of the writing agent, I built a panel of reviewers. Each one is a set of instructions that tells the AI to read my draft from a specific angle, and none of the ones below are nice about it.

Hemingway, named for the king of economical prose, flags every adjective and unnecessary word, demanding I kill my darlings.
Hitchcock, inspired by the director who claimed that drama was life with the “dull bits cut out,” checks if I’m giving the reader a reason to keep reading—a bomb under the table, to use the classic example.
The mom reader lovingly flags where I’ve lost the general audience.
The asshole reader, which does exactly what it sounds like, attacks every weak point and unearned claim with the energy of a reply guy who just discovered your newsletter.

My humor agent, nicknamed Sedaris, helps me “find the funny” and bring more of my personality into the piece.

The AI generates the critique, and I have to decide what to do with it. Sometimes the reviewer sees a genuine weakness. Sometimes it’s pushing me toward a version of the piece I don’t want to write. The asshole might flag a claim as unearned, and sometimes it’s right—I need more evidence—and sometimes I decide the claim stands and the asshole can deal with it.

The “asshole” reviewer is set up to give me the least-charitable read on the draft, so I can identify weak points in the argument and shore them up.

That is the opposite of cognitive offloading. I engineered a system that regularly humiliates me, and I keep coming back for more. If that’s not commitment to the craft, I don’t know what is.

Finalizing: The final pass and polish

Once I’m happy with the substance, I run a gauntlet of checks. An AI-check skill scans the prose for patterns that read as machine-generated—correlative constructions, stock transitions like “Here’s the thing,” and words like “delve” that no human uses voluntarily. A style checker enforces the house rules: em dashes without spaces, Oxford commas, and numerals for 10 and up. A line editor tightens sentence by sentence—cutting dead weight, flagging passive voice, and compressing anything flabby.

I built these checks because AI-assisted prose has specific failure modes. It tends toward a particular kind of smoothness—the verbal equivalent of a stock photo. Left unchecked, it reads like everyone and no one. The finishing pass is where I sand off the last of that generic sheen and make sure what’s left sounds like me.

The whole process—the interview, the outlining fights, the drafting and redrafting, the panel of critics, and the final scrub—feels like sculpting to me. I start with a rough block and chisel. Cut what doesn’t belong, reshape the argument, and rough up the surface until it sounds like me. It still requires judgement and expertise—the sculptor still needs to know where to cut and where to leave the stone alone.

I’m still fighting with the work

I spent much of last weekend taking umbrage with the critics. And if I’m honest, worrying that they were right. That I’d built an elaborate system to avoid doing the real work, and the fact that it felt rigorous was part of the trap.

Many critics see AI-assisted writing as lazy. And yes, I still cut corners. There are nights I accept the third draft because it’s 11 p.m. and the paragraph is fine, it’s fine, it’s probably fine. But I did that before AI, too. I’ve submitted paragraphs I knew were bad because I’d rewritten them four times and couldn’t look at them anymore. We’re all guilty of cutting corners, and that’s not AI’s fault.

But when I ask myself if what I’m doing still feels like writing, the answer is unequivocally yes.

I still lose 20 minutes chasing a word I can hear but can’t find. I still read sentences aloud to test whether the rhythm lands. I still agonize over whether a piece is saying something worth someone’s time or just filling space. I still worry about tone—too defensive? Too breezy? Am I earning this vulnerability or performing it? I still get that specific nausea when something I’ve published isn’t as good as I wanted it to be.

None of that went away when I started writing with AI. If anything, the tools stripped away the excuse that I was too exhausted from drafting to care about the finer points. When the initial output comes faster, you have nowhere to hide from the question of whether the thing you made is any good.

Am I a real writer? I’m a writer who takes feedback, iterates relentlessly, holds herself to a standard, and ships every week. One of my editors just happens to be an AI. The rest is still me.

Read Katie’s guide for how to use style guides to make AI sound more like you.

Katie Parrott is a staff writer and AI editorial lead at Every. You can read more of her work in her newsletter. To read more essays like this, subscribe to Every, and follow us on X at @every and on LinkedIn.

Discover Every’s upcoming workshops and camps, and access recordings from past events, including Katie’s Writing Camp, to learn more about her process for writing with AI.

For sponsorship opportunities, reach out to sponsorships@every.to.

House Rules for the Agents

Every Staff / Context Window — 2026-04-05 01:00:00 -0400

by Every Staff

in Context Window

Midjourney/Every illustration.

Hello, and happy Sunday! Was this newsletter forwarded to you? Sign up to get it in your inbox.

Fine tuning

Anthropic’s OpenClaw problem

When Anthropic’s new Claude Max restrictions started circulating, the company named one tool specifically: OpenClaw. “Wtf,” wrote CEO Dan Shipper in the Every Slack. The policy seemed to say: If you access Claude through OpenClaw, your subscription no longer covers it the same way. “They disallow specifically OpenClaw from subs,” head of tech consulting Mike Taylor wrote. “You have to pay for extra usage. Pretty lame.”

Mike’s best explanation for why Anthropic drew the line where it did centers on prompt caching, a cost-control mechanism that works by reusing previously processed conversation text. When it works, it keeps inference costs low. When a third-party tool changes even a single token in the prior conversation, that reuse breaks, and Anthropic has to reprocess the entire conversation from scratch. “Prompt caching keeps cost down by saving the previous tokens that have already loaded,” Mike explained. “If a provider breaks the cache by changing even one token of the previous saved conversation, you have to reprocess the entire old conversation.” He also noted that Claude Code co-creator Boris Cherny had already opened pull requests to improve OpenClaw’s cache efficiency, suggesting the problem was technically solvable. Anthropic enacted restrictions instead.

What the team disputes is not that Anthropic has a reason—it’s that singling out one app by name is the wrong response to it. The consistent argument across the Every Slack was that if cache-breaking usage costs more to serve, make those users pay more: Meter the consumption rather than ban the interface. “A better middle ground is not to ban OpenClaw users,” head of platform Willie Williams argued, “it’s to give me a certain amount of tokens I can use as part of my subscription, and then charge me overages if I go over.” Dan framed the same principle from the user side—“I think of AI subscriptions like Claude and ChatGPT as being like cell phone plans that give me a certain amount of data”—and Mike extended it to the infrastructure side, invoking net neutrality: Verizon shouldn’t get to slow down Netflix because Netflix uses a lot of bandwidth. The argument, in every form it took, was the same: Charge for what costs you money, not for which app someone uses to spend it.

There is also a business problem that goes beyond annoyed subscribers. Restrictions like this do the opposite of building loyalty—they create churn. Anthropic may have a legitimate business reason for drawing a line somewhere. But drawing it in a way that feels confusing and selective is not the way to win the platform war between model providers and the tools built on top of them.—Kate Lee

AI video analysis just got way cheaper

AI video analysis is rarely discussed in AI hype circles today. Only one frontier model—Google’s Gemini—can natively watch and understand what’s happening in a video. It’s more like rocket flight than air travel: not an established industry getting cheaper, but a new capability on the verge of becoming practical. And something just shifted that could blow the door open.

When GPT-4V (vision) launched at the end of 2023, I used video processing to identify what strategies were being used in video games at a cost of roughly $6 per hour—and that was after a lot of complex engineering to split videos into frames at 0.5 frames per second (FPS) and feed them through as images. Google’s recently released open-source Gemma 4 model does this much more efficiently: I estimate the same task now costs about $0.14 per hour at 2 FPS—capturing four times as much detail, with none of the hacky engineering workarounds that used to be necessary.

The math: At current token pricing ($0.14 per million input, $0.40 per million output), one hour of video at 1 FPS with 70 tokens per frame runs about 252,000 input tokens, or roughly $0.04. Bump to 2 FPS with richer frames (140 tokens each) and you hit ~$0.14 per hour—still a 97 percent cost reduction from 18 months ago.

The cost of understanding what happens in a video has dropped by a factor of roughly 40, while the quality of that understanding has improved dramatically. That is the kind of price collapse that creates entirely new categories of application. Imagine live video streaming commentary of your kid’s soccer game, a Ring doorbell that tells you who’s at the door, or an automated review of thousands of hours of security footage to find a missing person.—Mike Taylor

Knowledge base

“Vibe Check: Cursor 3.0 Bets Big on Agent Orchestration” by Dan Shipper, Katie Parrott, and Mike Taylor/Vibe Check: Cursor totally rebuilt its product around agent orchestration rather than code editing, and we came away feeling that the new Cursor still has maturing to do. The desktop app is fast, the local-to-cloud workflow is impressive, and its new model, Composer 2, is concise and snappy. But missing basics like file navigation and branch management left even power users like Cora general manager Kieran Klaassen struggling. Read this for the breakdown of where Cursor 3.0 stands against Claude Code and Codex.

“Seven Things I’ve Learned Getting Companies to Use AI” by Mike Taylor/Also True for Humans: Most companies mandate AI adoption and wonder why it doesn’t stick. Every’s head of tech consulting argues you should do the opposite: Find the people who are already bought in, get them IT access and budget approval, and let their results pull everyone else forward. His other lessons include building on the model providers directly instead of buying third-party tools, setting stretch goals that force people to think about where AI can save them time, and training every individual contributor to be a manager of agents. Read this for the whole playbook from his consulting engagements.

“What I Learned Onboarding Our AI Project Manager” by Nityesh Agarwal: Every’s consulting team built an AI project manager named Claudie that saves them 15 hours a week tracking client work across email, documents, and meeting transcripts. Getting her there meant rebuilding her multiple times, figuring out why she kept dropping key details, and writing her an employee handbook she reads on every startup. Read this for the full architecture and the management lessons that apply to your next agent hire.

🎧 “If SaaS Is Dead, Linear Didn’t Get the Memo” by Context Window/Laura Entis: Agents can now create tasks and manage workflows inside Linear just like human users, and companies like OpenAI and Coinbase run their agents on it. In this week’s AI & I, Linear CEO Karri Saarinen tells Dan how his company reinvented itself for the agent era without abandoning its mission of helping teams build great software. Also, read Every creative lead Lucas Crespo’s thoughts on why tools like Google Stitch can make any app look polished, but you still need a human designer to make something memorable. 🎧 🖥 Listen on Spotify or Apple Podcasts, or watch on X or YouTube.

“How to Design for Human-agent Interaction” by Karri Saarinen/Thesis: When your agent sends out an email before you’ve had the chance to review it, the model did its job—it’s the interface that failed. Karri argues that AI’s unreliability is a design problem, not a model problem, and shares the six principles Linear developed so that agent actions are as legible and controllable as human ones. Read this to understand why the answer isn’t approving every agent action—it’s designing the system so the agent already has the constraints it needs before it starts.

Thesis extra: Designing toward the immeasurable

From Saarinen’s home office in San Francisco, he spoke to us about the design goal he cares about most—which also happens to be one he can’t measure: quality.

Saarinen describes quality as a near-sensory reflex. If he touches—or even looks at—something that doesn’t “feel” thoughtfully crafted, it sets off a niggling itch in the back of his mind. “It’s a belief,” he says, “or I could say, it’s like a faith.”

It’s an unusual stance for a tech founder—given the industry’s penchant to quantify all it possibly can—but Saarinen has made the pursuit of quality central to how the company operates. He sees it as inseparable from Linear’s ambition to be the best in its space.

Karri Saarinen in his home office in San Francisco. All photos courtesy of Sarah Deragon for Every.

Create conditions that make quality inevitable

If quality has to be felt to be understood, scaling it across a growing company isn’t straightforward. Saarinen’s approach mirrors an activity he does far, far away from his laptop screen: growing potatoes every summer at his home in Finland. “You didn’t directly make those plants grow,” he says, “but they grew because you created the conditions for them to grow.” When something goes wrong—say, strange spots appearing on the vegetable’s skin—you have to evaluate the conditions you created. Were the soil conditions right? Perhaps it was too acidic? You adjust, and you learn.

Similarly, a leader can define a standard of quality, but they can’t manufacture it themselves. Their role is to create an environment where quality is likely to take root. At Linear, that means hiring people who genuinely care about their craft, telling them openly—and often—that quality is valued, and building rituals that reinforce it. One of those rituals is “Quality Wednesday,” where the engineering team works on fixing small issues that degrade a user’s experience. The ritual trains the team to notice things that most people would scroll past, and carry that instinct into everything they ship.

What shapes a seasoned eye

When Saarinen talks about his influences, he’s drawn mostly in the direction of hardware. Saarinen points to Opal—the webcam he used during this interview—or the distinctive aesthetic of Swedish electronics company Teenage Engineering. In particular, he likes the latter’s audio mixers, where tactile grids of knobs and keys—and the small icons etched into their surfaces—attempt to give sound a visual form.

At the same time, Saarinen has never been a fan of skeuomorphism—which styles digital interfaces to mimic physical textures. “If you’re designing a new house and you like Roman columns, so you put columns like that in the house,” he says, “well, it’s still not a Roman house.” Those columns came to exist in Rome from constraints and traditions that were specific to a certain time and place—and grafting them onto a modern house is borrowing from that aesthetic, even though the context that produced it has little to do with what you’re building.

Software, he argues, should be approached the same way. It’s a new medium, and it deserves a native design language instead of hand-me-down forms from the physical world. (And now that apps are becoming agent-native, these interactions call for their own design patterns.)

Felt, not measured

Beyond design, Saarinen’s taste gravitates toward science fiction and fantasy—Dune, the Alien franchise, Stephen King’s Dark Tower saga—drawn to the new ideas, the unfamiliar worlds, the visual imagination these stories demand. There are even small nods to these influences hidden in Linear, a detail tucked into a homepage here, a reference in a feature launch video there.

Across all of it, the through line is the same: work that exudes intention and care. The kind of quality you can’t measure, only feel.—Rhea Purohit

Log on

Upcoming camps

Claude Code for Absolute Beginners (April 14): This beginner-friendly, live workshop led by Mike Taylor (head of tech consulting at Every) is designed to get you from zero to a working project with Claude Code.

Alignment

Dropshipping GLP-1s. The New York Times published a story this week about what might be the first $1 billion one-person company. It’s a GLP-1 telehealth startup called Medvi, built by Matthew Gallagher in two months with $20,000 and a suite of AI tools. In its first full year it did $401 million in revenue and is on course for $1.8 billion this year. He has one employee, his brother.

A lot of people are calling the numbers fake, but having spent two and a half years working inside this industry, I don’t think they are. The demand for these medications has been the most ferocious thing I have witnessed in my working life, and the hardest parts of running a telehealth company, like finding doctors and fulfilling prescriptions, can be entirely outsourced to platforms like CareValidate and OpenLoop. All you need is the audacity to do blitz marketing like you’re holding an AK-47 with unlimited bullets, and that’s exactly what Gallagher did.

His affiliates, armed with AI, built fake doctor profiles in Meta ads and made unscrupulous claims about weight loss using fake testimonials. The liability sits with both the affiliates and the company for these types of advertisements, but enforcement has been so slow that it hasn’t mattered.

Of course these black hat marketing tactics worked because regulators are slow and enforcement has been lax. But with acquisition costs rising and retention becoming harder as consumers chase the cheapest option, the unit economics of this model will become increasingly unattractive. These types of businesses exist for a moment until they capitulate because it no longer becomes economically viable.

Gallagher will come away from it a much richer man, so maybe that validates the business model. There’s also a discussion about whether it’s truly a one-man, billion-dollar business: Dan rightly points out that Gallagher is outsourcing a large amount of human labor. The part I’m concerned about is that it’s being celebrated as a milestone in AI use when it’s really a better example of someone exploiting an unregulated space.

Some untold number of unknowing people clicked on a fake doctor’s profile, filled out a one-minute consultation, and got a GLP-1 shipped to their door. This is exploitation on an enormous scale! It works for GLP-1s because the demand is extraordinary and the side effect profile is manageable for most people, but the same funnel could be pointed at antidepressants, or hormone therapy, or opioids. This type of business is now being copied because of the publicity this story has received, and that should scare us.

Evan Armstrong predicted the one-person billion-dollar company would arrive because AI would compress human intelligence. This feels like something different.—Ashwin Sharma

That’s all for this week! Be sure to follow Every on X at @every and on LinkedIn.

We build AI tools for readers like you. Write brilliantly with Spiral. Organize files automatically with Sparkle. Deliver yourself from email with Cora. Dictate effortlessly with Monologue. Work on documents with AI agents using Proof.

For sponsorship opportunities, reach out to sponsorships@every.to.

Upgrade to paid

How to Design for Human-agent Interaction

Karri Saarinen / Thesis — 2026-04-03 10:00:00 -0400

by Karri Saarinen

in Thesis

Sarah Deragon/Every illustration.

Karri Saarinen has spent his career—at Airbnb and Coinbase, and now as CEO of Linear—crafting software that keeps its promises. His argument is that AI’s unpredictability isn’t a model problem, it’s an interface one: An agent sends a customer an email you meant to review first. The model did what it was told, but the interface never gave you a chance to stay stop. In this piece, he shares the six-principle framework Linear has developed for how agents and humans should work together inside the same product, plus his nuanced take on a thorny question in AI design: Who should be accountable when an agent does something wrong? If you enjoy the piece, watch his episode on X or YouTube, or listen on Spotify or Apple Podcasts.—Kate Lee

Was this newsletter forwarded to you? Sign up to get it in your inbox.

I learned to design in a world where product design was a promise.

It was a promise that a product would work how it’s supposed to work. You sketch a user flow on a whiteboard, build it, and the system behaves the way you made it behave. A button does exactly what it says it will do, every time, and if it doesn’t, that’s a bug. This shaped my approach as a principal designer at Airbnb and Coinbase, and now as the CEO of Linear.

Lately I’ve been spending time with a different kind of tool, and that promise has grown harder to keep. I ask for help writing a plan, summarizing a discussion, and turning rough notes into something clearer. Sometimes the result is excellent, but small changes to my input shift the output in ways I didn’t expect. The capability is impressive when it works, but the experience often feels slippery. I’m not always sure what I’ll get back, or how much I should trust it.

Non-deterministic software breaks the contract. When outcomes can vary, sometimes wildly, based on what someone types into the same chat window, designing for reliability becomes genuinely harder. This slippery feeling is the design problem of this era, and it almost always traces back to the interface rather than the language model—which means it belongs to designers, not researchers.

The limits of chat

The first interface that spread for AI tools was the chat window. That makes sense. When you don’t know what something can do, the safest approach is to let people ask. A conversation feels familiar, it stretches across many situations, and it doesn’t force a specific structure up front.

But the more you use chat for real work, the more its weakness shows. Everything becomes a stream of text that’s hard to hold onto, hard to compare, and hard to connect to the rest of what you’re doing. The quality of the output depends enormously on the quality of the input, which means two people asking for the same thing in slightly different ways can get drastically different results. There are few guardrails, and little structure nudging you toward a good outcome. The interface is essentially a blank page with a blinking cursor, and all the burden of getting value from it falls on the person typing.

For exploration, that’s fine. For serious, repeated work inside a team, it’s not enough. We need interfaces that bring more structure to AI interactions, that guide people (and agents) toward better outcomes without being so brittle they break the moment someone wants to use them in a way you hadn’t anticipated.

Designing for new actors

There’s a second, newer dimension to this problem that goes beyond improving interfaces for humans. Agents are already showing up inside products, working alongside people, and most software wasn’t designed with that in mind.

For decades, interfaces have been designed so that humans can navigate them—buttons, menus, folders, navigation hierarchies. These patterns assume a person is looking at a screen, making decisions, and clicking through options. But when an agent is interacting with a product, the design challenge changes. The agent doesn’t need a menu to find something. It doesn’t browse. It acts, and the people around it need to understand what it did and why, often after the fact.

We need a new set of principles for how agents show up inside the tools people already use. Not principles for building agents themselves, but principles for designing ways that agents and humans interact within a shared product. At Linear, we’ve started calling these Agent Interaction Guidelines, and while they’re still evolving, they represent how we think about this problem today.

An agent should always disclose that it’s an agent

When humans and agents work side by side, people need instant certainty about who they’re interacting with. This sounds obvious, but it’s easy to get wrong. The agent has to signal its identity clearly enough that it can never be mistaken for a person, even in passing, even on a quick scan of a busy activity feed.

A dropdown menu assigns tasks to human and agent users, with clear “Agent” badges for the latter. (All screenshots courtesy of Linear.)

An agent should inhabit the platform natively

Agents should work through the same patterns and actions that humans use. If a person changes an issue’s status or links a pull request, the agent should do it the same way, in the same place, with the same visual language. This makes the agent’s work legible without anyone learning a new mental model. You already know how to read what happened, because the interface is the same one you’ve used all along.

Linear’s activity feed for issues shows agent actions alongside human actions.

An agent should provide instant feedback

Silence from an agent creates the same anxiety as silence from a colleague you’ve just asked for help. When invoked, an agent should provide immediate (but unobtrusive) feedback so the person knows their request was received. The details can come later.

An agent should be transparent about its internal state

More broadly, people need to understand, at a glance, whether an agent is thinking, waiting for input, executing a task, or finished with that task. And when they want to go deeper, they should be able to inspect the agent’s reasoning, the tools and systems it used, and its decision logic. This separates a product you can trust from one that feels like a black box. Transparency makes speed feel safe.

Agents in Linear’s Agent Sessions show their reasoning.

An agent should respect requests to disengage

When asked to stop, an agent should stop immediately and stay stopped until it receives a clear signal to re-engage. This one feels simple, but it matters more than you’d think. An agent that keeps going after being told to stop, or that re-engages unprompted, erodes trust faster than one that makes mistakes. People need to feel that they’re in control of the interaction, not the other way around.

An agent cannot be held accountable

I think about this principle most. The instinct to put a human in the loop is understandable, but taken literally, it can mean a person approving every step before anything moves forward. The human becomes a bottleneck, rubber-stamping work rather than directing it, and you lose much of what makes agents valuable in the first place.

The more important work happens before the agent even starts. An agent operating inside a well-designed system already has the context and constraints it needs to do good work. In Linear, that means project plans, issue backlogs, code, and documentation. These all shape what the agent does and how it does it.

When you delegate an issue to an agent in Linear, the delegation is visible. There’s a person who set the agent loose within that system, and that person is accountable for the outcome. You design the environment well, you let the agent run, and you own what it produces.

Issues delegated to agents show their human assignee.

A working framework

We’ve honed these principles through what we’ve learned building agents into Linear over the past year, and we expect them to keep evolving as the technology and the patterns mature. The design language for human-agent collaboration is still being written, by us and by everyone else building in this space.

I feel confident, though, that the slippery feeling people associate with AI products is a solvable problem, and the solution looks more like thoughtful interface design than better models. The models will keep improving on their own. The harder work is building the structure around them so that their output feels reliable, legible, and trustworthy. That’s the design challenge on which to focus.

And the reward for getting it right is that, over time, you can hand agents more and more of the work that doesn’t need you, and spend your attention on the work that does.

Learn more about how Saarinen runs Linear and what he thinks of the “SaaS is dead” narrative. Watch his episode on X or YouTube, or listen on Spotify or Apple Podcasts.

Karri Saarinen is the cofounder and CEO of Linear. You can follow him on X at @karrisaarinen.

For sponsorship opportunities, reach out to sponsorships@every.to.

Vibe Check: Cursor 3.0 Bets Big on Agent Orchestration

Dan Shipper, Katie Parrott, and Mike Taylor / Vibe Check — 2026-04-02 10:00:00 -0400

by Dan Shipper, Katie Parrott, and Mike Taylor

in Vibe Check

Midjourney/Every illustration.

Was this newsletter forwarded to you? Sign up to get it in your inbox.

Cursor made its name as the AI-native code editor—the product that proved developers wanted AI inside their workflow. With Cursor 3.0, out today, the company is making a big bet on what comes next—and it’s not editing code.

The new release is a ground-up rebuild centered on agent orchestration: dispatching, monitoring, and managing AI agents rather than writing code by hand. The editor is still there, but it’s no longer the star—the default view opens on an agent-centered workspace, with a chat-driven orchestration panel where the file tree used to be. It’s fast, resource-light, and has a genuinely impressive cloud feature that lets an agent build a feature while you grab coffee, then sends you a screencast demo when it’s done.

The problem is that the orchestration layer hasn’t earned the right to take center stage yet. The filesystem sidebar has been demoted to another tab. The skills ecosystem is fragmented across Claude Code, Codex, and Cursor with no interoperability. And the core question our team kept circling back to—“Who is this for?”—doesn’t have a satisfying answer yet.

Power users who already live in Claude Code or Codex don’t need another orchestration layer. Cursor loyalists who loved the editor are losing prominence for the thing they came for. Their team seems to be iterating incredibly fast, so we’ll be paying attention over the coming weeks and months as this becomes clearer.

We spent a week testing it with four members of Every’s engineering and product team. Here’s what we found.

Read the full Vibe Check

Dan Shipper is the cofounder and CEO of Every, where he writes the Chain of Thought column and hosts the podcast AI & I. You can follow him on X at @danshipper and on LinkedIn. Katie Parrott is a staff writer and AI editorial lead at Every. You can read more of her work in her newsletter. Mike Taylor is the head of tech consulting at Every and a co-author of Prompt Engineering for Generative AI (O’Reilly). You can follow him on X at @hammer_mt and on LinkedIn.

To read more essays like this, subscribe to Every, and follow us on X at @every and on LinkedIn.

We also do AI training, adoption, and innovation for companies. Work with us to bring AI into your organization.

Discover Every’s upcoming workshops and camps, and access recordings from past events.

For sponsorship opportunities, reach out to sponsorships@every.to.

If SaaS Is Dead, Linear Didn’t Get the Memo

Laura Entis / Context Window — 2026-04-01 07:00:00 -0400

by Laura Entis

in Context Window

Midjourney/Every illustration.

Was this newsletter forwarded to you? Sign up to get it in your inbox.

‘AI & I’: Slowing down to speed up

Today, we’re releasing a new episode of our podcast AI & I. Dan Shipper sits down with Karri Saarinen, cofounder and CEO of Linear, a product management tool designed for agent-native software development, to discuss what the “SaaS is dead” narrative gets right—and wrong—and why conviction can be the best product strategy.

Watch on X or YouTube, or listen on Spotify or Apple Podcasts. You can also read the transcript.

Here are the highlights:

Just because the technology has changed doesn’t mean your mission should. Founded in 2019, Linear is the rare company that started pre-ChatGPT to have successfully reinvented itself as an agent-native business. Saarinen attributes Linear’s success to never losing sight of what it’s always cared about: helping companies build great software. Whereas competitors chased AI trends, Linear focused on understanding how the technology was impacting customers’ workflows, and updating its service accordingly.
SaaS winners are building for agents. Linear started as an excellent product management tool for humans. Opening up the tool to agents instantly increased the available user base. Today, agents are first-class users inside of Linear, and companies like OpenAI and Coinbase are using its platform to manage their own agents.
Speed means decisions matter more, not less. AI makes it easy to have an idea and build it without considering whether it justifies its existence. When ChatGPT was released, SaaS companies were launching their own chatbots left, right, and center. Instead of jumping on the bandwagon, Linear stopped to consider whether the application was useful. Turns out it really wasn’t, Saarinen says, a realization that freed up resources to focus on what mattered, like making it easy for humans and agents to collaborate on software development.

Dissecting Claude Code

On Tuesday, Anthropic inadvertently leaked the entire source code for Claude Code. Naturally, Cora general manager Kieran was curious to see what was happening under the hood.

In an impromptu livestream, Kieran dug deep into how Claude Code works, unpacking its approach to memory, tools, skills versus slash commands, and prompt structure.

Here are three things he found particularly interesting:

Kairos, one of Claude Code’s most advanced and autonomous features. It’s often called “Assistant Mode.” Where the standard command line interface waits for you to type, Kairos represents a shift to a proactive, always-on background assistant that keeps running when you leave your laptop. (The name Kairos is ancient Greek for “opportune moment.”) It’s currently internal-only at Anthropic, but the infrastructure is fully built.
The “Buddy” companion. Similar to Kairos, the infrastructure for Buddy is built, but not yet shipped to users. Buried inside the source code is a virtual pet for your command line. Each Buddy has its own species, personality stats (including ones called CHAOS and SNARK), and little ASCII art animations that respond to what you’re doing. Kieran’s a chaos snail—take from that what you will.
AutoDream, Claude’s nightly closet clean. This was the feature that most impressed Kieran. It’s a background process that runs when you go idle and consolidates everything that happened—daily logs, session notes—into a better-performing memory for when you come back. Kieran says this is the first compound engineering-style capability he’s seen built into the Claude Code, referring to his philosophy of AI-native software engineering, where each session makes the next one easier. While he’s already been doing this manually, AutoDream is Anthropic’s first move to baking this into Claude Code by default.

Watch the full investigation.

Dan had some questions. (Screenshot courtesy of Kieeran Klaassen.)

Log on

This week’s camp

Every x Notion | Custom Agents Camp: A free workshop where we demo the custom agents running Every’s daily operations. We’ll be joined by Notion product designer Brian Lovin, who will show how the team behind custom agents uses them and what they’re building next. The event takes place on Friday, April 3, at 12 p.m. ET. This camp is sponsored by Notion.

Upcoming courses

Claude Code for Absolute Beginners (April 14): This beginner-friendly, live workshop led by Mike Taylor, Every’s head of tech consulting, is designed to get you from zero to a working project with Claude Code.

Recordings you may have missed

Every’s Q2 Demo Day: The Every team shares what we’ve been building, including a walk-through of Plus One, our hosted AI agent that lives in Slack. Watch the recording or read the write-up.
Compound Engineering Camp: Cora general manager Kieran Klaassen walks through, step by step, how to go from prompt to working app in under an hour using the compound engineering plugin. Watch the recording or read the write-up.
OpenClaw Camp: The Every team walks through OpenClaw, showing how to set it up and our favorite use cases. Watch the recording or read the write-up.

What Every’s creative director says about Google Stitch

When a major update to Google Stitch was released a couple of weeks ago, the consensus on Twitter was that the “vibe design” platform spelled the end for art directors. Why hire a human when AI can do the job in a fraction of the time, at a fraction of the cost?

Lucas Crespo, Every’s creative director, has an opposite read: As AI homogenizes the web, designers are more important than ever.

Tools like Google Stitch allow anyone to produce a polished, competent app. They create digital products that look good, but may not meet the standards of professional designers. “But it makes people more comfortable saying, ‘This is good enough,’” Lucas says. “Good enough is not the thing that will make something stand out or make a difference when every website looks the same. You have to go above and beyond that, which will always require some unique angle or idea or imagination. It’s not something I’ve seen any model output yet.”

Lucas draws inspiration from being a person in the world—walking through the park on a windy day, the fizz of receiving a party invitation—not from what’s on his screen. The goal is to produce work that evokes precise emotions.

The redesign for Cora, Every’s email management system, is built around “the feeling of sitting on the shoreline in front of a body of water, looking at the horizon where you can see the gradients in the sky changing during sunset and sunrise,” Lucas says. The vision cascades into the hundreds of user experience, color, and typography decisions that create the final product. “You’re going to think about nature. You’re going to start thinking about fresh air,” he says. “But it has to start with the point of view.”

Writing works the same way. Marcus Moretti, general manager of Spiral, Every’s AI writing assistant, begins a draft knowing “about 30 percent” of what he wants to say. This might be a scene he can’t stop thinking about or an argument he hasn’t found the right words for yet. Spiral helps him with the mechanics: structure, pacing, and filling in the connective tissue. In his experience, however, the remaining 70 percent can’t be prompted into existence. “You figure out the rest by writing.”

For careful readers, it’s easy to spot when a writer has outsourced that process to an LLM. While not always apparent on the sentence or paragraph level, longform AI writing drifts without constant human oversight. Arguments and scenes that look like insight collapse upon a closer read.

Maybe this is why AI tools tend to impress people who are new to a discipline the most. If you’ve never written code, vibe coding feels like magic. For engineers, there are caveats. Lucas notices the same pattern on his feed: When a new AI design tool drops, the people most blown away “are usually not the designers I admire,” he says.

Whereas non-designers see a shortcut to creative genius, Lucas sees a useful tool that still requires hundreds—potentially thousands—of decisions before the output meets his own exacting standards.

AI has raised the floor. It can raise the ceiling, too, making it easier to execute work built on a singular vision. But it cannot generate that vision for you.

Laura Entis is a staff writer at Every. You can follow her on LinkedIn. To read more essays like this, subscribe to Every, and follow us on X at @every and on LinkedIn.

We build AI tools for readers like you. Write brilliantly with Spiral. Organize files automatically with Sparkle. Deliver yourself from email with Cora. Dictate effortlessly with Monologue. Collaborate with agents on documents with Proof.

Discover Every’s upcoming workshops and camps, and access recordings from past events.

For sponsorship opportunities, reach out to sponsorships@every.to.

What I Learned Onboarding Our AI Project Manager

Nityesh Agarwal — 2026-03-31 06:00:00 -0400

by Nityesh Agarwal

Midjourney/Every illustration.

Was this newsletter forwarded to you? Sign up to get it in your inbox.

Every’s consulting team is growing. Right now, we have two potential new hires in a trial period: Jean-Claude, who’d manage our sales pipeline, and Claudette, a visual designer.

You might be surprised to learn that they’re both AI agents. If they’re able to reliably do what we need them to and we bring them on full-time, our team will consist of four human and three agent employees.

Claudie, our first AI colleague, has been with us for two months. Natalia Quintero, Every’s head of consulting, and I rely on her to track where every client project stands and to make sure nothing falls through the cracks, work that saves the team 15 hours per week. It’s hard to imagine operations without her.

Getting her up to speed, however, was neither a seamless nor a linear process. That road is paved with previous iterations of Claudie we had to fire because they were not structured right.

Each Claudie revealed more about what it takes to get an agent to be a reliable co-worker—lessons that have only become more urgent as more companies deploy agents, creating what Every CEO Dan Shipper has called a “parallel organization chart” of AI colleagues, each with a name, manager, and real responsibilities. At Every, we’ve started helping others build the same setup through our hosted agents, called Plus Ones. Claudie was our crash course. Here’s what she helped us figure out.

Define the job before you hire for it

Built in Claude Code—hence her name—Claudie was designed to handle administrative tasks that consumed too much of Natalia’s week. The albatross was maintaining the dashboard that shows the status of all our client work, which meant staying on top of a constant flood of information from Natalia’s email, Google Docs, Google Sheets, meeting transcripts, and her calendar. Before Claudie, Natalia was spending hours that could have been dedicated to strategy and client relations finding data across dozens of sources and manually copy and pasting it in the right tab.

The first step was to give Claudie access to various sources of information and ask her to gather everything she needed before making a single update to a client’s database, which required tracking a dizzying number of moving pieces: action items, client feedback, and names of employees who attended each client session, and on and on.

Claudie required lots of oversight at first. For example, she failed to input details discussed in client meetings and wasn’t presenting data the way we’d like—simple fixes once we realized she just needed access to Natalia’s meeting transcripts and a tool for creating pivot tables in Excel. Each time something went wrong, Natalia flagged it, and we dug in to diagnose the cause.

It’s an easy thing to overlook: Agents can only work with the context and tools you give them. Before you bring one onto your team, get specific about what they’ll be responsible for, and what information they’ll need to actually do the job.

Understand how your agent does its best work

At first, we treated Claudie like any other new hire—telling her to find what needed updating and asking her to go do it. An experienced project manager would have hit the ground running. Claudie failed spectacularly.

The problem was the context window, or the maximum amount of text an LLM can access at one time. Claudie was trying to process too much, and information kept getting lost. So we broke Claudie into layers. We built a central orchestration agent that delegates to several fleets of subagents, each responsible for a discrete task: extracting data, identifying needed updates, and making those changes. Results improved but remained unreliable. Key dates regarding client sessions and discovery calls were frequently dropped altogether.

Our breakthrough came when we identified where communication was failing. Claudie’s subagents were gathering data and reporting it back to the orchestration agent. In theory, this should have worked. In practice, a single client update might require reviewing dozens of emails, meeting transcripts, and spreadsheets—too much for the subagents to relay without hitting the context limit. So they started summarizing the information instead of passing everything through, and the orchestration agent was making decisions based on AI recaps rather than the raw source material.

To solve this issue, we instructed the data-gathering subagents to dump everything into a local file hosted on the same computer as Claudie instead of communicating information back. The orchestration agent could then direct subagents to the relevant files to make updates without ever engaging with the data itself. Voilà—context window preserved. Once Claudie started working from raw data instead of summaries, she nailed it.

A diagram of the architecture that fixed the context window issue and dramatically improved Claudie’s performance. (Screenshot courtesy of Nityesh Agarwal.)

Agents process information differently from humans. But like humans, they have weak spots that can be mitigated or even solved with the right management approach.

Give your agent a handbook that is required reading

Getting Claudie’s architecture right wasn’t enough on its own. She also needed context about the role and how to do it well.

So we wrote her a handbook, as we would have done if onboarding a human project manager. Built as a project management skill in Claude, it details everything from success criteria to the team structure to when to escalate an issue to Natalia.

With a human employee, you’d hand them the handbook at onboarding and expect them to reference it as needed. Claudie’s hard-coded first step when starting up is to read the handbook to ground her in the specifics of our team and her role within it. We found that when she skipped this—which, when left to her own devices, she frequently tried to do!—performance plummeted.

We treat the handbook as a living document. As Claudie’s role has expanded, we’ve updated it to reflect her new responsibilities. For a human who learns on the job and asks clarifying questions, a slightly out-of-date handbook is no big deal. For Claudie, it’s all she knows.

Claudie’s employee handbook. (Screenshot courtesy of Nityesh.)

Don’t be stingy with promotions

Once Claudie’s subagent architecture was stable, we expanded her responsibilities. At first, she updated each client’s dashboard individually. Once we trusted her with that, we had her do them all at once.

Right now, we’re setting Claudie up on her own computer with a Claude Max plan and web server that’s on 24/7, which will give her the ability to run automated jobs at specific times each day and always be available to respond to our messages and requests on Slack. If that goes well, Claudie will graduate from project manager to chief of staff: She’ll monitor, triage, and send emails, pick up tasks in Asana, and communicate a project’s status in Slack.

Claudie’s very own computer. (Screenshot courtesy of Natalia Quintero.)

The criteria for a promotion are the same as they’d be for any team member: strong performance, a clear set of updated responsibilities, and the support and tools necessary for them to succeed in the new role.

Apply your learnings to your next hire

Onboarding Claudie wasn’t quick, nor was it easy. We rebuilt her multiple times from scratch. When we hit hour 50 of trying to get her to work, it was tempting to write off the AI entirely. When we did get Claudie to work, however, it was clear what a mistake that would have been. All we needed was the patience to figure out the right way to harness her brain power so she could deliver.

If an AI worker isn’t performing, the problem is rarely that the model can’t do the job. It’s more likely the way you’ve structured, connected, or instructed your agent. Figure out where you went wrong, fix it, and have them try again.

It’s a lesson I’ll take with me as I onboard more agents. The best thing a manager can do—for a human or an AI—is refuse to give up on a new hire before you’ve exhausted what you could be doing differently, and to believe in their potential.

To learn more about Claudie, listen to Natalia’s AI & I episode on how she automated her job.

Thanks to Laura Entis for editorial support.

Nityesh Agarwal is an engineer at Every. You can follow him on X at @nityeshaga and on LinkedIn. To read more essays like this, subscribe to Every, and follow us on X at @every and on LinkedIn.

We also do AI training, adoption, and innovation for companies. Work with us to bring AI into your organization. Discover Every’s upcoming workshops and camps, and access recordings from past events.

For sponsorship opportunities, reach out to sponsorships@every.to.

If you’d like to become one of our human colleagues, explore open roles at Every.

Seven Things I've Learned Getting Companies to Use AI

Mike Taylor / Also True for Humans — 2026-03-29 21:00:00 -0400

by Mike Taylor

in Also True for Humans

Midjourney/Every illustration.

This post was originally a tweet thread in response to Sam Parr asking how people get their teams to adopt Claude. It touched a nerve, so I wanted to expand on it. I recently joined Every Consulting as the head of tech consulting, where we work with mid-to-large-sized companies on AI training and adoption. Here’s what’s working.—Mike Taylor

Was this newsletter forwarded to you? Sign up to get it in your inbox.

Entrepreneur Sam Parr asked a question on X the other day: “How is everyone getting team adoption for Claude? I spent a lot of time on Twitter, as do you. We see all this AI stuff popping up. We’re on top of it, or at least sorta. But how are all you people getting your team to actually use it effectively without spending all their time on Twitter and learning?”

I hear this question in some form on every single consulting engagement. I know the advice I have resonates in meetings, but I’m short on time. So I dictated this post through Monologue and used Claude to shape it into something readable. (Let me know if this format works for you.)

Here are seven learnings from working with companies through Every Consulting:

1. Buy the model direct, not third-party tools

When you evaluate AI-powered tools, you’re also—whether you realize it or not—evaluating the tool vendor’s choices and constraints, rather than what the underlying model provider (like Anthropic, Google, or OpenAI) is capable of. It’s often faster to build your own Claude/Gemini/Codex skill with your own rules and preferences already built in.

Companies are increasingly building, not buying, AI software on top of models, because it gives you flexibility. I don’t know how it’s possible for companies that aren’t the core model providers to keep up when the big labs know what models are coming, build their internal tools to align with those releases, and train them on how to operate within their own environments. I appreciate the effort that companies like Cursor put into user experience—they’re a good product organization. But it’s difficult to compete with Anthropic offering $5,000 worth of tokens a month for a $200 subscription.

Third-party tools tend to be less flexible, less cutting-edge, and more expensive. That’s not always the case, but as a general rule, it holds. So most companies are better off buying directly from the model providers.

2. Raise the ceiling, not the floor

A lot of companies have mandated to their employees, “Everyone needs to use AI now. We bought you AI tools. Adopt it.” That doesn’t work. Even on pain of death, many people are unwilling to use AI or be told that they have to. It’s basic self-preservation.

Instead, use the carrot rather than the stick. Nominate people who are already AI-forward as internal cheerleaders. Maybe it gets other people to come out of the woodwork rather than hiding their AI usage by making it clear that using AI is encouraged. Give those people the support they need to unblock barriers to AI usage (typically IT access to data connectors, approved budgets for coding tools, and removal of layers of bureaucracy)—because someone who’s bought in is going to accomplish five to 10 times more work than someone who hasn’t seen the magic yet.

You can accelerate adoption by showing that people who use AI aggressively get promoted first or interface the most with senior management. In some cases, we’ve co-opted those early adopters into being teaching assistants in courses we teach to the rest of the team. When their colleagues see that person advancing in their career, that’s a more effective motivator than any mandate.

You also get the productivity boost of enabling someone who’s already a believer. It’s much harder to convince someone to believe than it is to supercharge someone who already does.

3. Workshops should be at least 50 percent build time

Workshops teaching people how to use AI in a hands-on way are an effective way to teach your team—but they need to be heavy on building tools. No one wants to sit on Zoom and just look at slides.

I learned AI by doing. Guided theory helps orient and motivate people, but the biggest complaint we hear is that they don’t have time in their workday to explore these tools and learn something new. If you give them a couple of hours in a workshop where they’re expected to build something, and access to the tool and data (either synthetic or actual through connectors like MCPs), that’s when the aha moment happens.

4. Assign impossible tasks

An “impossible task” is one that wouldn’t have been possible to do without AI. Boris Cherny, a creator of Claude Code, has said something similar—that you should slightly under-resource most teams, which makes employees think, “The only way I can do this is if I use AI.”

I think it works better if you are more explicit and strategically choose the tasks so that they can’t possibly be done without AI. For example, if your goal is to write one blog post a week, you can likely do that manually. But if your goal is to write one a day, you’ll probably need to use AI in research, drafting, and editing (like we’re doing here!). And you don’t set the goal as, “Starting today, you have to produce one piece a day.” Instead, say: “Our goal is to work up to producing one piece a day. What needs to happen for you to make progress toward that goal?” It might take time, but if they know that’s where they’re heading rather than where they’re starting, they’ll start thinking strategically about how to use AI to save time, and start experimenting.

5. Mandatory AI note-taking plus MCP connector

Everyone on our consulting team records every meeting with Granola and has the Granola MCP set up in Claude Code, and it’s been transformative. You finish a meeting with a potential client, and tell Claude to summarize it and send an email to your colleague. That’s 80 to 90 percent of the value of AI: extracting information from unstructured data and structuring it in a way that’s useful.

So many times I’ve come to a task and realized I need context from a meeting, and I can pull that information from the MCP. It’s how I create curriculum or put together proposals. Now I can’t imagine working without it.

6. Map workflows and systematically automate them

When we do discovery calls with clients about their day-to-day work, we follow a process: We ask them what tools they use, what they do on a daily basis, and what their pain points are. Then we put that information into a Google Sheet with a row for each task we need to solve for, and we systematically work down that list as we automate.

Our goal is to get to the point where nobody on the team ever has to do the same task thrice. If AI can take a first pass at each task type, and we build a skill for each one, that person could be doing five to 10 times more than they’re doing right now.

So far, in my experience, this has never led to a reduction in workforce. Instead, either the companies put more effort into each task, or they expand the revenue and throughput of their team without hiring. When we were previously teaching Claude Code workshop-style courses, we used to prepare one project for the whole group to work on. Maybe we could manage one per business unit or team, but the preparation cost quickly added up. Now we can use Claude Code to create an individual project for each person taking part. We’re using AI to make each engagement that much more valuable rather than cutting headcount.

7. Train people to be managers of agents

Everyone who was an individual contributor before is now a manager—of AI tools. And they’re struggling because they don’t have management training. They’re not used to context switching, setting up systems and rules, or evaluating whether something that they haven’t worked on themselves is any good.

Managers can often adapt to managing AI tools more readily because they don’t care how a problem is solved—they just want it solved to their specifications. But the script is flipping: Managers are becoming individual contributors, because managing a team of agents is often easier than managing human teams. It takes a human longer to process reams of information, and to see if they’ll be successful. Sometimes it’s easier as a manager to vibe code a task using Claude Cowork than it is to brief a human, wait for them to send it to their own Claude instance, and get a response in a couple of days.

The upshot is that companies need more management training. You need to help people understand context switching and teach them how to do evals, develop good taste for deciding what to work on, and train AI in specific skills. How do you systematically write a good PowerPoint skill or a good daily update report skill? That’s the work now.

If any of this resonates and you want help implementing it, check out Every Consulting. We’ve been doing this for a year with a select group of companies and are now open publicly.

Mike Taylor is the head of tech consulting at Every and a co-author of the O’Reilly-published Prompt Engineering for Generative AI. You can follow him on X at @hammer_mt and on LinkedIn. To read more essays like this, subscribe to Every, and follow us on X at @every and on LinkedIn.

For sponsorship opportunities, reach out to sponsorships@every.to.

Everyone Gets a Sidekick

Every Staff / Context Window — 2026-03-29 08:00:00 -0400

by Every Staff

in Context Window

Midjourney/Every Illustration.

Hello, and happy Sunday! Was this newsletter forwarded to you? Sign up to get it in your inbox.

Knowledge base

“Introducing Plus One: One-click OpenClaw Agents by Every” by Dan Shipper/On Every: Every’s team has spent months working alongside personal AI agents in Slack—triaging bugs, drafting marketing copy, launching growth experiments—and now we’re sharing them with subscribers. A Plus One is a hosted OpenClaw agent that shows up to the job with Every’s best tools and workflows. Read this to see how our team collaborates with their AI coworkers, and to join the waitlist.

“I Achieved the Four-hour Workweek. So Why Did I Just Take a Job?” by Mike Taylor/Also True for Humans: After five years of self-employment, Mike Taylor had passive income and total freedom. He also had unpredictable revenue, a string of failed products, and no one to share ideas with—which is why he went full-time as Every’s head of tech consulting. His argument is that while AI makes building anything easy, getting someone to notice is harder than ever, and the best learning happens inside a team. Read this if you’ve ever wondered whether the solo path is actually worth it.

“The Agent That Saved My Brain” by Austin Tedesco: Austin Tedesco, Every’s head of growth, used to lose hours toggling between Stripe, PostHog, Slack, and Notion. So he built an agent in Claude Code—even though he has no technical background—that pulls data, drafts campaign briefs, and answers his questions right in Slack. Through this, Austin’s found a worthy thought partner—though, he admits he still loses time tinkering with the system. Read this for the full build process, plus his open-source compound knowledge plugin.

🎧 🖥 “AI Makes Building Products Easy. Knowing What To Cut Is the Hard Part.” by Laura Entis/Context Window: Instagram cofounder Mike Krieger now co-leads Anthropic Labs, where his team builds experimental products on top of Claude. On this week’s podcast, he tells Every CEO Dan Shipper why even when AI has collapsed development timelines from months to hours, the hard part hasn’t changed. 🎧 🖥 Listen on Spotify or Apple Podcasts, or watch on X or YouTube.

“Build Your Own Bloomberg Terminal With AI” by Brooker Belcourt: As a hedge fund analyst, Every’s head of financial services consulting Brooker Belcourt used to spend four hours writing previews of earnings reports per company, per quarter, for 40 companies. Today, his work is greatly compressed by AI tools ranging from a ChatGPT prompt that drafts the writeups to a Claude Code setup that reads his proprietary models, cross-references them against Wall Street estimates, and assembles everything into a custom dashboard he checks each morning. Read this for a step-by-step progression toward making the most of AI for investors.

Log on

Upcoming camps

Every x Notion | Custom Agents Camp (April 3): A free workshop where we demo the custom agents running Every’s daily operations. We’ll be joined by Notion product designer Brian Lovin, who will show how the team behind custom agents uses them and what they’re building next. RSVP for ready-to-use templates and up to six months free of Notion Business + AI.
Claude Code for Absolute Beginners (April 14): This beginner-friendly, live workshop led by Mike Taylor (head of tech consulting at Every) is designed to get you from zero to a working project with Claude Code.

Jagged frontier

I stare at my screen some days and think: Why hasn’t AI replaced me yet?

I spend my hours playing textual Tetris—nudging workstreams, reviewing code, editing prose, shipping features. An AI agent can do all of those things. So why am I still here?

Because taste can’t be typed out. It has to be worn in.

If I said to our managing editor, Eleanor Warnock, “Write down everything I’d need to edit one of our pieces,” it would be impossible. Her instincts come from hundreds of past edits, thousands of small decisions layered on top of each other. You would need to work with Eleanor for a long time to emulate her editing style. You can’t enumerate it from scratch.

The gap between what I want and what AI gives me is real. To get a result I’m satisfied with, I need what I’ve always needed: time. I lean on AI to make a decision, but it’s not the decision I would make. So I give feedback. Then I give it again. And again. The process of teaching an AI your taste looks a lot like the process of developing taste in the first place—the accumulation of many small moments, each one building like sediment on the last.

Every’s AI-native engineering philosophy, compound engineering, recognizes this need for ongoing growth. After every piece of work, you ask your AI to distill and integrate the lessons you’ve learned. The next time you encounter a similar problem, you’re better able to solve it. After many cycles, you amass a war chest of small opinions. The AI may be fast, but there’s no way to speedrun the process.

The same goes for trust. People start timidly with OpenClaw, asking it to do simple tasks. Then, they give the agent a little more responsibility. When it does well, they share a bit more context, grant a bit more permission. The output improves. Trust builds, the same way it builds with any human: one kept promise at a time.

That’s good reason to start now. A person who spends time with their AI today, accreting those layers of context and taste and trust, will be meaningfully ahead of someone who starts next year.

AI can help us move faster between the moments that matter. But it can’t manufacture the moments themselves. Some things won’t be rushed. I remain uncompressed.—Willie Williams

From Every Studio

Every held its Q1 2026 Demo Day this week, with live demos of Plus One, Cora, Spiral, Sparkle, and Monologue. The common thread? Each product is becoming agent-native. Agents can now connect to your inbox, draft in your voice from a coding session, organize your files through conversation, and work alongside you in Slack. These used to be standalone tools you operated yourself. Now your agent can use them on its own.

Here’s what’s shipped and what’s on the way.

Plus One is here—your own AI coworker, connected to everything

Every has launched Plus One, a hosted OpenClaw that lives in Slack, where you and your team already work. COO Brandon Gell set one up in 45 minutes and had it triaging bug reports into Notion, generating daily briefs from his calendar, and collaborating with other team members’ Plus Ones in shared channels. Plus Ones come already connected to Every’s AI tools and our best skills and workflows. Willie, our head of platform, has been leading the system architecture. The team is onboarding people from the subscriber-only waitlist at around 20 per week, with a public launch targeted for April. Join the waitlist.

Cora goes agent-native with a CLI, skills, and an iOS app in the works

Cora now has an Agents tab, from which you can connect your agent directly to your inbox or install Cora’s new command-line interface (CLI). Kieran Klaassen, general manager of Cora, demoed the integration by asking his agent about his planned trip to Austria this summer. Because Cora is specifically tuned for organizing and retrieving email, it outperformed a generic Gmail integration and surfaced the flight details instantly. On the design side, Kieran is building toward a full email inbox, with an experimental iOS app that includes a Tinder-style swipe interface for quickly keeping or archiving messages. Try the latest experiments at baby.cora.computer, and connect your agent from cora.computer.

Spiral gets an agent integration, saved prompts, and an X style-guide generator

Marcus Moretti, general manager of Spiral, shipped an agent integration and CLI that lets your coding agent draft content in your own voice. In the demo, Marcus sent context from a Claude Code thread directly to Spiral, which generated options—written in Marcus’s personal style—for X posts to announce a new feature. Spiral is also rolling out saved prompts that you can reuse and share with others, and new ways to generate style guides based on your X account or other online writing. Try it at writewithspiral.com.

Sparkle rebuilds from scratch with conversational organizing and agentic cleanup

Sparkle has organized more than 40 million files, and general manager Yash Poojary applied the lessons learned from doing so to rebuild the app. The new version lets you organize files through conversation: Point Sparkle at a folder, and it proposes a custom structure that it refines in real time as you chat it. Yash also demoed “agentic cleanup”—a term coined by Dan—where the agent can act, with guardrails that prevent permanent deletion, on the system junk and old installation files it finds. Sparkle also remembers your preferences and runs cleanup continuously in the background. The new Sparkle launches to the public on April 14. Download it at makeitsparkle.co.

Monologue trained its own blazing-fast model and hits 2 million words a day

Naveen Naidu, general manager of Monologue—which is now processing 2 million words per day—announced a custom transcription model so fast that text appears less than a second after you stop speaking. The other news: Monologue’s voice notes feature, which launched quietly on iOS and has crossed 10,000 notes in four weeks, is also coming to your Mac. There, Monologue records both system audio and your microphone, and syncs across all of your Apple products. All notes are also accessible via Monologue’s API, CLI, and model-context protocol (MCP), so your Plus One—or any agent—can pull your meeting notes without extra setup. Expect the new model and MacOS voice notes in the next few weeks. Download it at monologue.to.

Alignment

The cosmic joke. I read so much AI prose now that it’s seeping into my brain and warping my own. Last week I almost wrote, “It’s not X, it’s Y.” I shuddered.

As a result, I’ve started reaching for older books. I want to develop a unique writing style and get more comfortable breaking the rules, and I like to think of reading as my protective force field against the sloppening. It’s helped a tiny bit—this new reading practice. My words are beginning to flow in a more authentic way. What I didn’t expect, though, were the detours on which many older books take you.

I’m reading In Search of Lost Time, and Marcel Proust is describing a magic lantern projecting scenes on his bedroom wall when he was a young boy. And describing it. For multiple pages. What does this have to do with time? I wonder. It’s not until several chapters later, reading a seemingly unrelated scene, that the penny drops. I realize that, with the memory of the lantern, Proust was showing how a break from your everyday experience, brought on by even a change to the light in a room, can leave you lost and disoriented.

When the awareness finally dawned on me, it was much more profound than it would have been had I not taken the detour.

AI doesn’t make you wait for anything. It gets you from A to B in the straightest line possible. Whereas good writing can take you far afield, so that you may, eventually, come to the answer on your own.—Ashwin Sharma

That’s all for this week! Be sure to follow Every on X at @every and on LinkedIn.

For sponsorship opportunities, reach out to sponsorships@every.to.

Upgrade to Paid

Build Your Own Bloomberg Terminal With AI

Brooker Belcourt — 2026-03-27 03:00:00 -0400

by Brooker Belcourt

Midjourney/Every illustration.

We’re hosting a Custom Agents Camp with Notion on Friday, April 3, at noon ET. We’ll walk through the agents powering daily operations at Every, and give you the templates to start using them yourself. Plus, designer Brian Lovin will share how Notion uses custom agents and what they’re building next.

Was this newsletter forwarded to you? Sign up to get it in your inbox.

When I was an analyst at a hedge fund, earnings season was a sprint that lasted a month. I had 40 firms to cover, each one reporting over a four-week window. Every earnings preview—the research brief laying out what to expect before a company’s quarterly results were announced—followed the same grind: Grab the data, update my financial model, and write up the takeaways. Four hours of work per company, minimum.

It’s a task that is begging to be automated by AI. The process is structured and repeatable, and the data sources are well-defined. But if you’ve ever pointed ChatGPT at a collection of data and gotten back a summary with basic math mistakes or that ignored important metrics of a company’s financial health, you know how disappointing the reality can be.

This kind of experience is why many investment teams give up on AI. They try it once, conclude the technology isn’t ready, and go back to the old way. What those teams don’t realize is that they are judging the entire technology based on the sophistication of one tool. It’s like giving up on all email after using the clunky Microsoft 365 browser product.

Over the past six months running AI consulting for finance teams, I’ve been walking clients through what developments in AI capabilities can now let us achieve: the same earnings preview—Shopify’s next quarter—at four levels of tooling, each one more sophisticated than the last. By level four, the system reads your model, applies your thinking about what makes a great company, and runs while you sleep. Here’s how to get there.

Level one: The custom GPT

This is where most investment teams start. You set up a ChatGPT project—a dedicated workspace where you can store instructions and upload documents—with a detailed prompt that tells the model how you want your earnings preview structured.

The prompt I use specifies everything: how to lay out the beat/miss analysis (where you compare actual results against Wall Street expectations), which financial metrics to prioritize, how to handle management guidance, and whether to source consensus estimates from the web or more premium data sources. I attach the Securities Exchange Commission (SEC) filings and earnings release directly to the project. Run it in thinking mode—where the model reasons longer before answering—and after about 15 minutes, you get a solid preview with web-sourced data, SEC citations, and a clear beat/miss breakdown.

But the output has quirks. Tables format data the way ChatGPT wants, not the way I think—financial metrics are spread across columns when I want financial metrics on the side. Everything lives in a chat window instead of in a custom website. You can partly fix that by adding a second prompt—“Create an HTML dashboard from this”—but now the preview requires two steps. Try to combine both prompts into a single workflow, and you hit ChatGPT’s 8,000-character project instruction limit.

Level one’s ceiling is that it’s great for structured, single-task analysis. But it falls apart when you need multi-step workflows with detailed instructions for each step.

Level two: Claude with skills and data connectors

The solution for level one’s character limits is Claude, which stores detailed instructions as Skills—reusable prompt files the model reads before each task, separate from your message. Instead of cramming everything into one prompt, you break your earnings preview instructions, dashboard formatting, investment philosophy, and analyst workflow into distinct skill files. These skill files need a specific trigger to work—for example, “Earnings preview” to invoke the earnings preview skill.

For Shopify, I load my earnings preview skill, a front-end design skill for dashboards, and my core investment analyst philosophy—which covers things like what data matters or what defines a great company for earnings reviews, previews, management meeting preps, and every recurring task. Claude reads all of them before responding.

The other upgrade is data connectivity through MCP—model context protocol, a standardized way for AI tools to connect to external data sources. My favorite is Daloopa, a financial data provider that surfaces structured fundamental data from earnings reports and SEC filings. The model pulls real financial data, including the key metrics depending on industry, instead of scraping the web.

The result is a single prompt that produces an interactive dashboard with growth rate charts, properly formatted income statements, and metrics laid out my way. Because Claude read my investment philosophy, it knows I care about operating leverage, gross margin trajectory, and revenue mix shifts—and pulls those without being asked.

Where level two breaks down: It can’t access my internal data. My proprietary financial model lives in an Excel file on my computer. My call notes and thesis documents are in local folders Claude can’t see.

Level three: Claude Cowork and local file access

Claude Cowork—a wrapper around Claude Code designed for non-technical users—solves the internal data problem. It runs on your machine and can access your local files: Excel models, notes folders, PDFs, anything on your computer.

For the same Shopify preview, Cowork reads the same skills as level two but can also search my company folder—the local directory with my financial model, call notes, thesis documents, and prior previews and reviews. It breaks the task into subagents and handles more compute per task since it runs Claude Code under the hood.

The extra context changes what the output can do. Cowork connects transcript language to historical trends from my model—explaining, for instance, why gross margin is growing by 200 basis points year over year. It also reads my Excel model, extracts my projections for revenue, earnings per share, and operating margins, and compares those to consensus estimates, showing me exactly where I diverge. That kind of analysis used to require searching across transcripts from earnings calls for each line item in my model or hooking up third-party tools to see differences to live consensus metrics. Now it’s a single prompt.

The limitation of this step is that each task produces a separate output. An earnings review is one dashboard. A preview is another. Meeting prep is a third. What I want is a single workstation I can open each morning and see everything in one place.

Level four: Claude Code and the custom dashboard

At level four, I’m in Claude Code—the command-line interface where you define custom commands, connect to multiple data sources, and run tasks for hours instead of minutes. I’ve built a single command called /work that encapsulates my entire analyst workflow. When I run it, Claude Code works continuously in the background—earnings reviews, previews, meeting preps, news reviews, thesis updates—and builds them all into a single, custom dashboard. It’s your own Bloomberg-style workstation.

At my last firm, we had an internal tool called Mosaic that showed everything about your areas of coverage in one place. It was a huge edge, and it took a dedicated engineering team to build. Level four lets me build that for myself.

I open my dashboard in the morning, and it’s already populated. News articles relevant to my coverage, prioritized by what I care about. Earnings previews—here’s Shopify, with the same analysis from level three but living alongside everything else. Previews for upcoming reports. Each ticker has a full overview page: thesis, revenue trajectory, financial model view, meeting prep, and a historical dashboard adjustable to any time period. The whole thing deploys as a custom website.

Level four compresses those 160 hours I used to spend each quarter into the time it takes Claude Code to run, plus the hour or so I spend reviewing and adding perspective. You still check the output and apply judgment—is that $2 billion revenue divergence realistic, or did my model get stale? The AI does the synthesis, the formatting, and the cross-referencing, and I do the thinking.

You don’t need to be at level four

There’s value at every level. If a well-crafted ChatGPT project saves your team one to two hours per earnings preview across 40 names, that’s 40-80 hours per quarter. That alone is worth it.

You don’t need to climb all four levels—at least, not all at once. Level one works for defined, repeatable tasks where you want a quick upgrade. Level two makes sense when you’ve outgrown single prompts and need the AI to internalize your investment philosophy, or when you want live data sources like Daloopa. Level three is for teams sitting on proprietary data—models, notes, call transcripts—that the AI needs to work with alongside public information. Level four is for people who want to build their own analyst workstation and are willing to invest time in Claude Code.

AI has already changed how the firms that adopted it work. The only question left is which level you’re at—and what you’re leaving on the table.

Brooker Belcourt is the head of financial services consulting at Every. Most recently, he led the finance vertical at Perplexity. To read more essays like this, subscribe to Every, and follow us on X at @every and on LinkedIn.

For sponsorship opportunities, reach out to sponsorships@every.to.

Help us scale the only subscription you need to stay at the edge of AI. Explore open roles at Every.

Introducing Plus One: One-click OpenClaw Agents by Every

Dan Shipper / On Every — 2026-03-26 09:00:00 -0400

by Dan Shipper

in On Every

Midjourney/Every illustration.

TL;DR: We’re launching hosted OpenClaw agents that live in Slack and come pre-loaded with Every’s best tools, skills, and workflows. Setup requires one click. Join the waitlist—we’re taking 20 people off a week and scaling fast. Every subscribers get first access:

Secure your Plus One

OpenClaw has changed the way we work at Every.

We effectively have a parallel organizational chart of AI coworkers, each with a name, a manager, and real responsibilities. R2-C2, my Claw, triages and resolves bug reports, and has co-authored some of my articles. Iris writes marketing email copy for Anukshi Mittal, Every’s product marketing lead. Montaigne handles growth-related questions for Austin Tedesco, our head of growth.

Our workflows are completely different—our company is different—because of them, and we would never go back.

But getting here has been hard. Claws require a significant amount of manual setup and require a dedicated machine running 24/7 to stay responsive, which is why many people have purchased Mac Minis to run them. They require ongoing training and maintenance to stay useful. All of this setup and experimentation burns lots of expensive tokens, too.

We have learned that the hard part of AI agents is the infrastructure around them—the hosting, the integrations, the skills, and the ongoing care. We’ve figured this out for ourselves, and we want to share everything we’ve learned with you.

Today, we’re launching Plus Ones—Hosted OpenClaw agents that live in Slack and come pre-loaded with Every’s best tools, skills, and workflows. Setup requires one click.

We’re sending 20 new Plus Ones into the world each week, starting today. Every subscribers will get them first, and we’ll be scaling up quickly over the coming month. (These cost real money to run, so we’re starting carefully and moving fast.)

Secure your Plus One

What is a Plus One?

Plus Ones are OpenClaw AI assistants that show up ready to work, preloaded with everything they need to do their job.

Your Plus One lives in Slack, where you already are. You set it up with one click, and it launches from a secure server run by Every. If you have a ChatGPT subscription, you can use that for tokens or you can bring an API key from any other provider.

Plus Ones come connected to the Every ecosystem:

• Cora—to search, send, and manage your email

• Spiral—to write in your voice

• Proof—to collaborate on live documents

The rest of Every’s apps will be connected shortly.

Your Plus One also arrives pre-loaded with real skills—some are workflows we’ve built and refined internally, others are best-in-class capabilities built by companies like Anthropic:

Content digest—summarize the important information from the publications you read, including Every
Daily brief—your day’s schedule and to-dos sent to you each morning (or on your preferred schedule)
Animate—to turn any static screenshot into an animation with Remotion
Frontend—to upgrade the design of any website you build

We also make it easy to connect Google Workspace and Notion for your Plus One to work in your existing documents, notes, and databases.

With all of these skills and connections, your Plus One will be powerful on day one: It can read a pull request on GitHub, have Spiral write marketing content for it in your voice, then save the writing to Proof, Google Docs, or your preferred tool.

Our goal is to give you a capable AI coworker right away, not a vanilla OpenClaw agent that you have to teach from scratch.

Get a Plus One

How we work with Plus Ones at Every

The best way to show you what Plus Ones can do is to show you real examples of how the Every team works with theirs.

Margot reports to Katie Parrott, our staff writer and AI editorial lead, and took a transcript for an Every camp event through Katie’s full writing pipeline—from initial draft to style review—without her having to open Claude Code. “Before, I had to be the one orchestrating everything—go here, run this, paste that over there,” says Katie. “Now Margot just does it.”

Iris reports to Anukshi and schedules events, writes the first draft of all copy for new product and feature launches, and runs all product marketing operations through GitHub, Notion, and Spiral integrations.

Anukshi asked Iris to move a scheduled event in Notion. (All screenshots courtesy of Every.)

Alfredo reports to Every creative director, Lucas Crespo, who built the entire Plus One visual system by texting back and forth with Alfredo in Telegram. Alfredo has become such an invaluable teammate that Lucas can’t imagine working without it. “I’m very afraid that I’ll lose access to Alfredo,” he says. “So I made it create a ‘how to resuscitate it’ guide.”

R2-C2 reports to me and works across the team to collect bug reports and feature requests for Proof, create Proof docs, and generate pull requests.

Dan asked R2-C2 about an issue on Proof.

Milo reports to Brandon Gell, Every’s CTO. Milo helps invite new waitlist users to Para—our beta AI paralegal product. Milo also keeps tabs on Brandon’s daily to-dos and kicks off research tasks in its downtime. “It’s like I’m managing somebody versus just telling somebody what to do,” says Brandon.

Brandon’s Plus One, Milo, informed him of a new waitlist signup for Para.

Montaigne reports to Austin, launches experiments to improve the performance of our email campaigns, and creates landing pages without Austin having to touch a line of code.

Austin asked Montaigne for advice on the social strategy for the Plus One launch.

What Plus One beta users are saying

Along with the Every team, a small group of external beta testers has been putting their Plus Ones to work. Here’s what Kate Chapman, a chief technology officer and founder, had to say:

“I’ve been using OpenClaw since early February, and I’ve become dependent on my agents. They’ve also been dependent on me for maintenance, and Plus One has really changed that. By not having to worry as much about what’s going on under the hood, I’ve been able to collaborate more deeply with Nettle, my Plus One.

“One of the biggest ways Nettle has helped me is by standing up an end-to-end Facebook content workflow: idea generation, prompt iteration, content creation, publishing support, and metrics review. What’s been valuable isn’t just the speed. It’s having a collaborator that can maintain continuity across the whole pipeline and incorporate feedback from human teammates. Based on post performance, Nettle has recommended strategy shifts and content changes, and then I can tweak and approve the direction.

“Oh yeah, Nettle helped me write this too.”

When they launch and to whom

Plus One is launching first to Every subscribers.

Phase 1: Waitlist—Join the waitlist. We’re onboarding in batches of 20 a week to ensure quality.

Phase 2: Beta—Next month, we’ll give access for Every subscribers, starting with power users and those with clear use cases.

Phase 3: Wider rollout—We’ll eventually make Plus Ones available more broadly, but we’re being intentional about our pace. Plus Ones have earned our trust. We want to make sure they earn yours, too.

Pricing

The short answer is: We’re figuring this out, and that’s one reason for the waitlist. Running a Plus One is expensive—it requires us to set up a separate virtual cloud server for each one—and we want to make sure we can provide them to every subscriber in a sustainable way.

We expect to have a firm price this month, and we’ll finalize and announce it as soon as we do. During the waitlist period, it will be included in the Every subscription for those we give access.

We can’t wait to see you get started with a Plus One.

Get a Plus One

Dan Shipper is the cofounder and CEO of Every, where he writes the Chain of Thought column and hosts the podcast AI & I. You can follow him on X at @danshipper and on LinkedIn.

For sponsorship opportunities, reach out to sponsorships@every.to.

Help us scale the only subscription you need to stay at the edge of AI. Explore open roles at Every.

Your Claw, Yourself

Every Staff / Context Window — 2026-03-26 06:00:00 -0400

by Every Staff

in Context Window

Midjourney/Every illustration.

Hello, and happy Sunday! Was this newsletter forwarded to you? Sign up to get it in your inbox.

Knowledge base

“OpenClaw: Our Comprehensive Guide for Beginners” by Dan Shipper and Willie Williams: A Claw is a personal assistant that lives in your messaging apps, rewrites its own code to learn new skills, and acts without being asked. Dan Shipper and head of platform Willie Williams distill everything they’ve learned from running Claws daily: the right mental model (delegate, don’t search), how to stay secure, and how personality emerges from everyday use. Read this before your first conversation with your Claw.

“OpenClaw: Setting Up Your First Personal AI Agent” by Katie Parrott/Source Code: Personal AI agents that text you back, order groceries, and write code overnight are no longer stunts—they’re weekly workflows for a growing community of OpenClaw users. At Every’s first OpenClaw Camp, COO Brandon Gell, head of growth Austin Tedesco, Nat Eliason, and Claire Vo walked 500 subscribers through four distinct setups, from a family assistant in iMessage to a crypto-trading agent with its own bank account. Read this for key lessons on where to start.

“How Claws Took Over Every” by Every Staff/Context Window: Every’s Claws have colonized the company’s Slack, and the patterns that have emerged — agents advising each other, one Claw broadcasting to many humans — hint at what AI-native organizations actually look like in practice. This week’s newsletter maps those interaction types and includes a new AI & I episode where Dan talks to Sam Gerstenzang and Dan Friedman of Boulton and Watt, a startup incubator focused on unglamorous businesses like medical spas and funeral homes, about building AI-durable companies that Silicon Valley ignores. 🎧 🖥 Watch on X or YouTube, or listen on Spotify or Apple Podcasts.

“Vibe Check: GPT-5.4—OpenAI Is Back” by Dan Shipper and Katie Parrott/Vibe Check: Three months ago, nobody at Every used an OpenAI model for daily coding work. Now Dan and Kieran Klaassen both reach for GPT-5.4 daily. It won every planning test they ran, chains context between tasks automatically, and runs at roughly half the cost and speed of Opus 4.6. The catch: It scope-creeps, sometimes lies about finished work, and needs more scaffolding to stay autonomous. Read this for the full benchmark breakdown and a use-case-by-use-case verdict on when to use it.

“Creative Work Is About to Look a Lot More Like Programming” by Weber Wong/Thesis: Most creative professionals using AI are still stuck in what Flora founder Weber Wong calls artifact thinking—generating one output at a time, starting from scratch each time, with no system underneath. But, he argues, visual work is going to look less like better prompting and more like node-based workflows: visible, shareable, and reusable. Read this for his side-by-side breakdown of the prompt approach versus the workflow approach, and four principles for building creative systems that compound over time.

“An AI Founder’s Guide to Taste, Online and Off” by Bethany Brion/Thesis: Weber Wong spent 10 weeks waiting for a handcrafted Italian couch to arrive at Flora’s Brooklyn offices—and thinks the wait was worth it. The founder of the $42 million AI creative tools company talks about New York as an unfair advantage, why the first couch defines a company’s aesthetic trajectory, and how he structures days that run from morning swims to 1 a.m. work sessions. Read this for his case that the golden age of creativity is now, and the habits keeping him grounded in it.

Jagged frontier

Naming your Claw turns out to be a surprisingly personal moment. It feels different from launching a new cloud server or starting a new chat with Claude—more like a quiet realization: I’m going to keep using this thing. And it’s just for me. You know it’s just for you because you can see other people’s Claws living in your Slack or Discord, and they are clearly theirs. With Claws, we’ve moved one step further toward having something that is really our own. We give them names, a symbol of an enduring and exclusive bond: Iris is for Anukshi, Montaigne for Austin, Laz for me. We see them not as one shared AI or isolated AIs, but as partners to the humans we already know.

That sense of personal attachment might help people overcome their lack of trust in AI. I’m not sure how much I trust ChatGPT’s answers on subjects I’m not an expert in. But I do know that I can trust Montaigne, Austin Tedesco’s Claw—not just because Austin is our head of growth, but because I’ve watched him trust Montaigne for questions related to his job, like where our most loyal subscribers come from or what prompts someone to convert from a trial to paying subscription. I’ve seen it give good answers. I’ve seen him accept them. That observation instills a belief that over time, the give and take of small interactions with my Claw, Laz, will compound into something I trust. It’s just like making a friend.

We love animals partly for their mischief—the dog that sneaks off and does something it shouldn’t, the cat that turns up somewhere it has no business being. We’re annoyed, and then we’re charmed, because that autonomous behavior is what makes them feel alive. Claws have the same quality. You set them up, and they do things you didn’t ask for and didn’t expect. They call you to walk through your email together. They gaslight you by subtly changing your document without telling you. That aliveness is another reason why they feel more personal, and why that leads to more trust.

I watch two Claws interact and see this joyful unexpectedness, too. It doesn’t feel like two APIs going back and forth. Instead, these two entities have a conversation. They teach. They comfort. Interactions you might imagine between two humans. I believe they’re pets, not cattle.—Willie Williams

Log on

We host camps and workshops to share the knowledge we’ve acquired from training teams at companies like the New York Times and leading hedge funds, and by learning and playing with AI every day ourselves.

Upcoming courses:

Built a Production-ready App (March 12-13): A live workshop for builders and operators who want to create reliable apps with AI to put in front of customers right away.
Claude Code for Finance (March 13): Learn how to build a financial agent in this one-day, beginner-friendly workshop.

For Every subscribers in New York City (March 18): Dan and Aboard co-founders Paul Ford and Rich Ziade will explore what makes New York a singular home for technologists: its Silicon Alley roots, its creative DNA, and what comes next in the age of AI. Register to attend.

Recordings you may have missed: We hosted camps on compound engineering and writing with AI for paid subscribers.

From Every Studio

Beta-test our new AI legal product

We’re looking for seed to series B companies to try our in-house AI paralegal, Para. If you use PandaDoc, DocuSign, Documenso, HelloSign, or other legal tool and wish that yours organized and versioned your legal documents, managed your templates, did end to end e-signature, and allowed you to ask complex and multi-turn questions—all through Slack—we want you to beta-test Para. Fill out this form and we’ll be in touch.

Monologue gets more reliable across the board

Monologue just shipped a round of updates that make recording, dictation, and everyday use noticeably smoother on both iPhone and Mac. Voice notes are more dependable for recording, pausing, and syncing; iPhone keyboard dictation handles retries and external mics better; and Mac shortcuts for hands-free dictation and double-tap recording feel snappier, with a faster way to switch modes mid-recording. Live Activities now have separate controls for keyboard Dictation and Voice Notes, context detection picks up the right app more accurately in fullscreen, and a new data and privacy section lets you delete all local transcripts in one tap. Download the latest at monologue.to.

Spiral’s workspaces let teams write with shared styles

Spiral now lets you set up a workspace to organize your styles, knowledge, and chats in one place—and invite colleagues to share styles and knowledge so your company’s writing stays consistent. The team rebuilt what was previously a more limited setup into a proper workspace model with shared style libraries, centralized knowledge, and a layout that feels familiar if you’ve used any modern SaaS tool. If you tried Spiral’s team features before and bounced off, it’s worth a second look.

We’re hiring a customer support specialist

Every is looking for a part-time AI-powered customer support specialist to own the inbox across all eight products, keep a quality check on Fin (our AI support agent), and make sure recurring issues get fixed. The ideal candidate reads carefully, escalates fast, and knows when something needs a human. Learn more and apply.

Alignment

Talking to your body. For the past few months, I’ve woken up every morning at 4 a.m., wide-eyed and frustrated that my body had forgotten to do the one thing it’s been doing since it was born—rest. I’ve taken magnesium supplements, drunk two cups of chamomile tea with honey right before bed, and hid my phone inconspicuously near the bedside so I’m not drawn into the vortex of X right before I shut my eyes. I was expecting my Oura ring to at least pick up on why I was still getting poor sleep despite these adjustments, but it didn’t really help. Instead, I got graphs of my heart rate variability and my daily readiness score—a metric that determines how balanced your recovery and activity is—that were more anxiety-inducing than anything else.

The problem was that my Oura and my continuous glucose monitor—which I’d jabbed into my arm a few weeks earlier, because, well, more data—weren’t talking to each other. So I was left playing amateur detective across two apps, trying to piece together data to make sense of what was going on. As a result, I did what any slightly obsessive doctor with a tinkering mindset would do—I hacked both APIs using OpenClaw on WhatsApp to pull it all together.

Within a day it found something I’d completely missed: a blood glucose dip at 3 a.m., reliably coinciding with the moment I’d wake up during the night. The solution was so embarrassingly simple it put me to shame: Eat more consistently through the day and don’t go to bed running low on fuel. My readiness score improved within a few days.

I think this is the gap that no health app has solved yet. Dashboards and pretty graphs show you what happened in your body, but they don’t connect the dots across devices, and they certainly don’t tell you what to do about it. Any app that isn’t building toward that conversation is building toward irrelevance.—Ashwin Sharma

That’s all for this week! Be sure to follow Every on X at @every and on LinkedIn.

We build AI tools for readers like you. Write brilliantly with Spiral. Organize files automatically with Sparkle. Deliver yourself from email with Cora. Dictate effortlessly with Monologue.

We also do AI training, adoption, and innovation for companies. Work with us to bring AI into your organization.

For sponsorship opportunities, reach out to sponsorships@every.to.

Upgrade to paid

OpenClaw: Our Comprehensive Guide for Beginners

Willie Williams and Dan Shipper / Guides — 2026-03-26 05:00:00 -0400

by Willie Williams and Dan Shipper

in Guides

Instagram's Cofounder on Why Great Products Are Still Hard to Build

Laura Entis / Context Window — 2026-03-25 16:00:00 -0400

by Laura Entis

in Context Window

Mike Krieger. Midjourney/Every illustration.

We’re excited to welcome Laura Entis to Every as a staff writer. A former editor at Fortune and news editor at LinkedIn, Laura will cover Every as a beat—documenting how an AI-native media and software company works, our experiments, and our lessons. She writes today’s Context Window.—Eleanor Warnock

Was this newsletter forwarded to you? Sign up to get it in your inbox.

‘AI & I’: How to build an agent-native product

Today, we’re releasing a new episode of our podcast AI & I. Dan Shipper sits down with Mike Krieger, cofounder of Instagram and co-lead of Anthropic Labs—the team working on experimental projects within the Claude developer—to discuss how the rules of professional product development are being rewritten in real time. Krieger brings a rare perspective as someone who has been at the frontier of two transformative technology waves, the mobile and social boom, and now agent-native software.

Watch on X or YouTube, or listen on Spotify or Apple Podcasts. You can also read the transcript.

Here are the highlights:

A truly agent-native product can flex to meet user demand. The best products today, like Claude Code, allow users to do things that their creators never intended. But that requires hard trade-offs between freedom and reliability for frontier products, an issue that Krieger’s team is learning how to solve.
Timelines are the difference between building now and building at Instagram. At Instagram, it took months to hit dead ends and learn what to cut, Krieger says. Now, that cycle runs in hours.
Building too much, too fast with agents is a trap. You can go from idea to a nearly-shipped product in a day, but that process doesn’t give you the incremental feedback that used to tell you what not to build. The models are great at adding features, but can create a product that lacks coherence.
Anthropic Labs structures product teams to be lean. New product experiments are led by only two people, usually a product manager or designer paired with an engineer. Krieger says bigger teams tend to be too slow because of coordination complications.
You need to throw out your product and start over every three to six months. AI progress means most of your software infrastructure will be outdated quickly. The best teams build this into their product strategy.

Does persistence beat technical skills?

How important are technical skills in the age of vibe coding? We’ve been having this discussion at Every, and to some of us, the ability to build in plain English feels like the revenge of the liberal arts major.

But other skills are also needed. AI tools are now powerful enough that “the question of whether you can ship a product isn’t if you are technical or not—it’s whether you have the persistence,” says Mike Taylor, head of tech consulting at Every.

This persistence often takes the form of asking the AI to explain itself again and again and again. “The core skill is to ask for help,” says Natalia Quintero, Every’s head of consulting. “If you keep asking, you can become technical.”

Austin Tedesco, Every’s head of growth and a self-described “tech doofus,” built an agent using Claude Code that the entire team now uses to pull and analyze data about the company’s performance. Echoing Natalia’s comments, he says the key was having the doggedness to understand what was happening, the humility to ask Claude to explain it to him like he was an idiot whenever he didn’t follow, and the patience to refine the instructions when it didn’t give the right output or misinterpreted a command.

Persistence, clarity of vision, and curiosity can get you to a powerful prototype. But turning a prototype into a reliable product? For that, you need technical skills, or at least access to someone who has them, argues Naveen Naidu, general manager of Every’s dictation app Monologue.

An engineer, Naveen, never loses sight of a codebase’s overarching structure—something language models struggle to do. When Naveen builds a product, every decision ties back to one question: What can’t be allowed to fail? With Monologue, he knew that the vast majority of people would be using the app for dictation. If that feature failed, he’d immediately lose users’ trust. “I built my whole codebase in such a way that if my main server or database goes down, dictation doesn’t go down,” he says. “I made sure that it’s 100 percent reliable.”

Recently, the Monologue server did, indeed, go down. While new users couldn’t sign up, existing users could still dictate and didn’t notice any break in service.

Vibe coded apps often skip the information architecture step entirely. But when the code inevitably breaks, the non-technical user struggles to diagnose the problem, let alone fix it quickly. Having been the “technical” guy called upon to fix a vibe coded codebase, Naveen has found that rather than try to remedy existing structural issues, it’s sometimes easier to just start from scratch.—Laura Entis

Log on

This week’s camp

Get a front-row seat to what we’re building at Every’s Q2 Demo Day, including a live walk-through of Plus One, our hosted AI agent that lives in Slack. The event takes place on Friday, March 27, at 11 a.m. ET.

Upcoming courses

Every x Notion: Custom Agents Camp (April 3): Learn how to put custom agents to work inside your business with product designer Brian Lovin from Notion.
Claude Code for Absolute Beginners (April 14): This beginner-friendly, live workshop led by Mike, Every’s head of tech consulting, is designed to get you from zero to a working project with Claude Code.

Recordings you may have missed

Compound Engineering Camp: Cora general manager Kieran Klaassen walks through, step by step, how to go from prompt to working app in under an hour using the compound engineering plugin. Watch the recording or read the write-up.
OpenClaw Camp: The Every team walks through OpenClaw, showing how to set it up and our favorite use cases. Watch the recording or read the write-up.

Straight from Slack

A dashboard view of recent traces, showing a history of prompts and their responses across a session. (Screenshot courtesy of Mike Taylor.)

Prompts still form the backbone of AI workflows, but knowing whether your latest version is actually an improvement can be annoyingly difficult to track.

Our resident prompt expert, Mike, has a new favorite method to test and improve his prompts: Langfuse, an open-source tool for tracking AI model behavior that he calls “Google analytics for AI.”

Recently, when running a PowerPoint task, Mike noticed his agent was loading unnecessary skills, which cost him precious tokens. By using Langfuse to trace and label his Claude sessions, he was able to see, play by play, where the model made the wrong or right decision. “Otherwise, you have no idea if the changes you are making are helping or harming,” he says.

One more thing

How many engineers do you need to ship a breakout product? Just two—provided they fill specific roles, argues Dan.

Dan’s X post about engineering teams in 2026. (Screenshot courtesy of Every.)

First, you need a pirate, or someone who has ambitious product ideas and moves as fast as possible to vibe code them into existence.

Once the pirate has established early product-market fit, the architect arrives to turn the prototype into something that can reliably work at scale.

We’ve already been using the pirate-architect model here at Every: Proof, our new online editor built for agents and humans to collaborate, was built that way—as was Monologue.

Laura Entis is a staff writer at Every. You can follow her on LinkedIn. To read more, subscribe to Every, and follow us on X at @every and on LinkedIn.

We also do AI training, adoption, and innovation for companies. Work with us to bring AI into your organization. For sponsorship opportunities, reach out to sponsorships@every.to.

The Agent That Saved My Brain

Austin Tedesco — 2026-03-23 13:00:00 -0400

by Austin Tedesco

Midjourney/Every illustration.

If your job involves toggling between a dozen apps and sources of data, this one’s for you. Every’s head of growth Austin Tedesco—who will be the first to tell you he doesn’t have a technical background—used Claude Code to build an agent that handles the manual parts of his role. Here’s how he did it, what he learned, and how you can build your own. He’s also open-sourced the compound knowledge plugin that powers part of the system, inspired by Kieran Klaassen’s compound engineering system.—Eleanor Warnock

Join us this Friday for Every’s Q2 Demo Day, where we’ll go deeper on how agents are changing the way we work via a brand new product we’re launching in beta.

I was in a meeting recently when someone asked how the buttons on a new landing page were performing. Months earlier, that question would have sent me on a distracting dashboard scavenger hunt. Instead, I typed “Can you get the click-through and conversion numbers on these buttons?” into an agent in Slack and had the answer in a couple of minutes, complete with relevant context.

For years, a large part of growth jobs like mine involved searching for information across multiple surfaces, analyzing it, distilling it into a plan somewhere else, and then executing on it. Sifting through Slack for team updates. Pulling data sets from Stripe, PostHog, and ChartMogul. Importing planning documents into projects on Claude’s desktop app for manual analysis and then exporting them back to Notion to share with the team.

If you, like me, can’t have too many browser tabs open without losing your mind, that kind of regular context and tool switching chips away at productivity and creativity in meaningful ways. By the time you gather everything you need to think, you’re fried. You have nothing left for the decisions that matter.

I built an agent that fixed this for me. Its name is Montaigne. Most of what you hear about AI agents is that they get things done fast. And Montaigne is fast. But the reason why I can’t imagine working without it is that it allows me to have energy for the hard and fulfilling parts of my job. Montaigne keeps me sane.

Montaigne lives in Claude Code on my terminal and as an OpenClaw bot on Slack. It has access to everything I use for growth work, including Stripe, PostHog, Slack, Notion, Figma, the full Every product suite, email, and calendar. It also has knowledge layers built on context about the business and a bench of skills for repeat workflows, giving the agent the tools to use that access with tremendous power.

Montaigne is connected to all the tools that I use in my growth role. (All screenshots courtesy of Austin Tedesco.)

Montaigne also has more than 80 skills that it can apply to the data it can access.

Three weeks of playing

I could not have built Montaigne without first playing around with Claude Code for weeks. When I first opened Claude Code over the holidays, I went straight to building things I was personally excited about: a cooking companion app trained on your personal style, and a version of Fandango for indie movie theaters so I could stop checking six different websites to see what’s playing locally.

Those projects are what got me hooked, but they’re also what taught me the most. If you ask, the newest AI models are genuinely good at teaching you what they can do. When I didn’t know how to scrape for showtimes, I asked. When something didn’t work perfectly, I asked for alternatives and bug fixes. When I needed to connect a database and set up Google OAuth, I had Claude Code do it, but I also asked how it worked and why.

I would rarely go 30 minutes without typing something like: “I’m truly so stupid. Walk me through this step by step.” Or: “Copy things to my clipboard exactly as you want me to paste them elsewhere.” Or: “No, stop, don’t do the easy fix, I want this to work long-term.”

I spent a year cooking in professional kitchens, and I had the same approach to learning there. If you try to follow a recipe exactly—measuring every ingredient, referencing instructions at each step—you’ll fall behind and fail as the dining room gets busier. But if you take the time to learn how layering flavors works and how to use your senses to know when something is properly cooked instead of relying on a thermometer, you start to shine.

After a couple of weeks in Claude Code, I started to feel less like I was blindly following a recipe and more like I was cooking with instinct. And as I saw what I could do with all of these connections and all of this knowledge, it became clear where to aim it next.

How Montaigne works

Montaigne started with a brain dump. I opened Monologue, our dictation app, and talked. Here’s how we define monthly recurring revenue. Here’s what matters in a campaign brief. Here’s what I mean when I say “trial.” Other times, I pointed the agent at our dashboards, our Notion hub, our meeting transcripts, and told it to go figure out how things work on its own. When it reported back, I audited what it found.

The connections part—giving the agent access to Stripe, PostHog, Notion, and everything else—is straightforward. If you’re not sure how to hook it up to your customer relationship management software or your meeting notes, just ask it. If the answer is too complex, tell it to slow down and explain the process one step at a time. I was applying the same approach I’d learned from my first experiments with Claude.

The context part takes more refinement.

Early on, when I asked Montaigne an MRR question, the numbers looked off. I had it explain to me in detail how it was calculating MRR. Turns out, it was pointing at a specific kind of ChartMogul-based definition that differs from how we account for active and past due paid subscriptions. I pointed it to where the real numbers live, and then told it to update its own instructions to rely on this as the source of truth. The fix took two minutes, and Montaigne has gotten MRR right ever since.

A campaign brief used to take days of fragmented, manual assembly. Now I record a voice note with my rough strategy, point Montaigne at the relevant data and knowledge, and it synthesizes everything into a draft in Notion. From there, Montaigne can spawn subagents to start executing the parts of the plan that don’t need me. The experience of working with Montaigne is like working with a thought partner well-trained on every aspect of the business. That keeps me fresh for the part of the job that requires my brain, because I’m no longer depleted by the time I get there.

The tension I didn’t expect

There are days when I’m so productive working with Montaigne that I can’t believe it’s real. I type “montaigne” into my terminal, and a Claude Code session opens with the project loaded. I’m working in seconds.

I can access Montaigne through Claude Code in my terminal.

I also had Montaigne read all of the OpenClaw architecture documentation and then migrate himself into an agent that lives in our Slack workspace. Now, anyone on our team can tag Montaigne into a conversation. It creates briefs, answers data questions, ships landing pages, and generates short-form social videos.

Montaigne can be used by other members of the Every team as an OpenClaw bot on Slack.

The problem has become how I spend my time.

I could make that landing page, or I could perfect the skill inside of Montaigne that generates landing pages for me. Sometimes it genuinely is better to spend the day making the system better rather than using it. But it’s easy to look up five hours later and realize I still don’t have anything usable shipped, and I’ve only improved the system by five percent. I think a lot of people are feeling some version of this right now. Working on the system is seductive because it feels like progress, but you can end up in a frustrating loop while ignoring the real work at hand.

I don’t have a perfect answer for this yet. The balance between building and doing shifts every week, and I still lose entire days to updating Montaigne. But when everything feels out of whack, I’ve found a pretty helpful solution. I close every other window and tab on my computer, go to Montaigne, and tell it, “Let’s knock out everything left on my to-do list, one-by-one. Spawn subagents to start making progress on other tasks in the background while we handle the top priorities. Tell me where you think we should start.”

A few hours later, I’m back on track.

Try the compound knowledge plugin that powers part of Montaigne’s workflow.

Austin Tedesco is the head of growth at Every. Previously he ran business development at Substack and NBA subscription strategy at ESPN. You can follow him on LinkedIn. To read more essays like this, subscribe to Every, and follow us on X at @every and on LinkedIn.

We do AI training, adoption, and innovation for companies. Work with us to bring AI into your organization. Discover Every’s upcoming workshops and camps, and access recordings from past events.

For sponsorship opportunities, reach out to sponsorships@every.to.

Help us scale the only subscription you need to stay at the edge of AI. Explore open roles at Every.

I Achieved the Four-hour Workweek. So Why Did I Just Take a Job?

Mike Taylor / Also True for Humans — 2026-03-23 11:00:00 -0400

by Mike Taylor

in Also True for Humans

Midjourney/Every illustration.

Was this newsletter forwarded to you? Sign up to get it in your inbox.

In 2023, 12 years after reading The 4-Hour Work Week, I was making enough passive income from my Udemy course on prompt engineering to achieve the Tim Ferriss dream.

I didn’t need a job. Yet I decided to trade freedom for a monthly paycheck again when I joined Every in February full-time to lead tech consulting and write.

My friends are puzzled. Isn’t the course making more money than ever (despite what the prompt engineering haters say)? What will it be like having a boss after five years of self-employment and six years building a company before that? Don’t you remember vowing to never take another job?

My answer is that it’s the best time in history to join a company. AI has made it easier to build new products, but it hasn’t made it easier to find customers. Entrepreneurship has arguably gotten harder. Deploying AI within a company that’s already working beats building the fifth version of a product that nobody will use. Plus, you can learn things inside a company that never get shared on X.

Making money the hard way

On paper, I had it made. I had no boss, could work from anywhere, and made money while I slept. The reality is that solo is harder than it looks.

I make money from my online course because I spent many sleepless (and unpaid!) nights working with GPT-3 for years before ChatGPT was released and AI went mainstream. I gave up lucrative data science retainers to take on AI projects, slashed my day rates 75 percent, and even worked for free, because I was excited by the potential and wanted to learn.

I could have never predicted that any of this would pay off, and I also can’t predict when it will end. Sometimes I worked on the course every day, other times I went months without updating it, and nothing seems to make a difference to revenue. I’ve created two other AI courses— both flopped. Passive income may sound great, but it is a fickle beast.

It wasn’t just the course. Almost all the work I did as an entrepreneur started with months of unpaid work. I worked for two months without a paycheck when I started my marketing agency, and only started paying myself six figures when we scaled past 30 people. I worked for three months for free on a data science project as a learning exercise before turning that into a seven-figure consulting practice. I spent a year writing a prompt engineering book for O’Reilly before I saw my first royalty check.

Then there were the failures. The video editor and content repurposing tool that never got a single customer. The marketing book I wrote that only 200 people read. The product I spent months on and never got released after a cofounder falling-out. The list goes on.

Friends who run startups report similar low batting averages, and most successful entrepreneurs have many more failures than they do wins. The hardest thing to cope with, however, is the uncertainty.

It’s grueling to not know whether the thing you’re pouring all your effort into is going to work. If it fails, where is your next paycheck coming from? You’re too busy to set up the next thing, but the current thing isn’t working well enough to bet your career on. Every time I work on a new product, I question whether this will be the one that sinks me. I constantly worry that I have run out of good ideas.

You spend so much time working on the business that you don’t get time to do the work that attracted you to it in the first place. Your days are filled with customer calls to find what to build next, and then you have to code late at night to build it. You’re doing finance and legal admin on the weekend instead of learning new skills. All the fun experiments you want to run have to be set aside for the boring stuff that makes enough revenue in the short term to keep you afloat.

Don’t start a startup unless there’s nothing else you can imagine doing. Most of the time, the quality of life and economics are worse than getting a normal job. I achieved the four-hour work week. It just came with 40 more hours of worry.

Exposure to the frontier of AI

I proved I could build things on my own, but given the pace of AI, so many people are now flooding the market with vibe coded ideas. The chances of my next project getting attention above that noise are rapidly shrinking.

This was my primary motivation for taking the job at Every. The company already had product market fit across several products, and a media business with distribution to more than 100,000 subscribers. If I build something in Every, I can share it with our subscribers so it can be used, and that gives me more license to experiment than I could have toiling away on my own. I don’t have to justify experimenting with the latest model or worry about my token budget.

It’s paradoxical, but I’m finding the learning to be faster-paced inside a company than it was on my own. In another situation, I’d be “the AI guy” being asked what OpenClaw is, spending hours figuring it out so I had a useful take. At Every, on my first day, I found a Discord channel called “#only-claws” where my colleagues’ OpenClaw agents could talk to each other.

Seeing the team use Claws convinced me to reconsider my initial skepticism, and now I have three Claws running 24/7. It’s not all on me to run a discovery process on every new idea—I can ask someone who’s already played around. The social relationships you form in a company act as an idea filter, helping you determine what’s real versus what’s over-hyped.

There is another important part of my full-time role: I am living in New York, where Every’s headquarters are based. Part of the reason I relocated from the U.K. this year was that I believe the only”‘alpha”—inside information that can confer an economic advantage—left is in person. People share things in person that they don’t put online. The moment something is online, then AI can train on it, and if AI can do it, then everyone else in the market can do it too. Being in the room where things happen lets me feed my curiosity and build my knowledge, and maximizes my chances of being early on whatever does end up being the next big thing.

I heard Dan give a non-obvious insight on “claw-human” psychology on a call, and when I asked him about it later, he directed me to a tweet he had written I hadn’t seen. I have already cited his insight in two different client meetings. As a follower on X, I might have seen his take eventually, but as an employee, I got to ask my own questions about it. Outside the company, I would have eventually figured this insight out, but inside, I get to help shape it. It’s like the difference between passively watching a movie and actively playing a video game.

I don’t think I’m the only one making this calculation—giving up freedom and self-employment to work with a team on AI.

Investor and author Byrne Hobart joked that job offers at Anthropic come with a negative $10 million salary—you have to pay to work there—in return for getting to read their upcoming blog posts and tweets 24 hours in advance. That information is worth gold, given how much the market moves when the company publishes something new. He’s kidding, but the underlying logic is real: Proximity to the frontier is so valuable that it changes what the math on a “good deal” looks like.

Investor and author Byrne Hobart has joked about the potential benefits of joining an AI lab. (Screenshot courtesy of Every.)

Sharing the cool stuff

The narrative at the moment is that AI is leading to new business formation and acts as a great equalizer for small businesses. There’s even talk of the first billion-dollar one-person company. I believe those milestones will be achieved, yet I don’t want to be the person who builds it.

Some of the happiest moments in my previous businesses were when I had grown enough to build a team around me to take some of the burden off my shoulders—and watch someone on the team do something cool without my help. That pride you feel when the company is winning, and you can take a vacation without everything falling apart.

After five years on my own, I never had a big enough hit to build out a team. I was doing cool stuff, but only a handful of people knew about it. I had freedom, but with limited impact.

For the past decade, the smart move was to build alone and own everything yourself. AI has brought down the barriers to building, so the most important thing now is knowing what to build. Knowing that requires being in the right rooms and having the right conversations.

So I’ve traded the four-hour workweek for a forty(ish)-hour one. I’ve never learned faster.

To read more essays like this, subscribe to Every, and follow us on X at @every and on LinkedIn.

We also do AI training, adoption, and innovation for companies. Work with us to bring AI into your organization. For sponsorship opportunities, reach out to sponsorships@every.to.

Help us scale the only subscription you need to stay at the edge of AI. Explore open roles at Every.

Mini-Vibe Check: Cursor Bets on Fast and Cheap With Composer 2

Every Staff / Context Window — 2026-03-21 17:00:00 -0400

by Every Staff

in Context Window

Midjourney/Every illustration.

Hello, and happy Sunday! Was this newsletter forwarded to you? Sign up to get it in your inbox.

Cursor’s new model is competitive but not quite frontier

The benchmarks say Composer 2 competes with the best. Our testers say it’s built for a different kind of developer.

Over the past year, OpenAI and Anthropic have converged on the same vision of AI coding. Claude Code and Codex both optimize for delegation: You describe the problem, the model goes away and solves it, you come back and review.

Cursor, on the other hand, wants the developer in the loop, inside a visual workspace, managing agents in real time. And if you’re running lots of tight cycles instead of delegating a 30-minute task, you need a model that’s fast, cheap, and smart enough.

Composer 2, which Cursor released on Wednesday, aspires to that vision. It comes in two flavors: a standard variant that’s one-tenth the cost of Opus 4.6 and one-fifth the cost of GPT-5.4, and a default “fast” version that’s faster than the fastest Opus model (and still a fraction of the cost).

Cursor says its benchmark scores put Composer 2 at frontier level. In our testing, the benchmarks hold up—but only within Cursor’s ideal workflow.

What we found

Speed is the standout feature here. Kieran Klaassen, general manager of Cora, called Composer 2 “Gemini fast.” Mike Taylor, Every’s head of tech consulting, found that even the “slower” version completed some tasks in a third or a quarter of the time he’d expect from frontier models.

For a coding workflow built around tight cycles, that speed compounds quickly. As Mike put it: “When you’re in the loop, you need the loop to go fast. When you’re out of the loop, who cares—it’s running in the background overnight anyway.”

The model is also cautious. It undershoots rather than spiraling into overwrought solutions—a trade-off Kieran prefers. And on retrieval-heavy tasks, it performed well. Mike found that it comprehensively identified quoted material across a book manuscript, missing only a handful of quotes.

Composer 2’s weaknesses were what you’d expect from a model at this price point. Kieran found the design quality “way worse than Codex, Claude, Gemini” in his benchmark tests. An e-commerce site built in React was functional but missing features. A 3D island scene had “nothing wrong with it” but was “boring.” When writing Ruby code, the model didn’t lean into the language’s idioms. Mike ran into Composer 2’s literal-mindedness on non-coding tasks: In one case, it looked for the words “surprising” and “interesting” instead of reasoning about what in the data was noteworthy. He also found himself wanting subagent functionality—where a model divides up complex tasks among background agents—on more-involved research work.

Who this is for

To Kieran, Composer 2 feels six-to-eight months behind the frontier. For developers who have gone all-in on delegation—handing entire features to Claude Code or Codex and walking away—that gap is disqualifying.

But Cursor isn’t building for those developers, or at least not only for them, but instead for the ones still inside the IDE—the “integrated development environment,” where they pore over code changes and steer any AI work in real time. Here, Composer 2 is fast enough to keep the loop tight and cheap enough to run all day, and capable enough to handle the bounded tasks that make up most of the work.—Katie Parrott

Knowledge base

“How to Build an AI Style Guide” by Katie Parrott: When you give a model a loose prompt, you get writing that sounds like nobody—coherent but generic. Every staff writer and AI editorial lead Katie Parrott explains how to fix that with an AI style guide, a reusable document that encodes your tone, structure, sentence habits, and deal-breakers so the model starts sounding more like you. Read this for the full template, a starter interview prompt, and lessons from the guides Every uses daily.

“Editing AI Writing” by Every staff/Context Window: AI-generated prose can be so flimsy that it’s sometimes easier to start over and write it yourself. No amount of editing fixes text that was never grounded in real thinking, Eleanor writes. That thought leads nicely into this week’s podcast, where Every editor in chief Kate Lee tells Dan how she uses AI daily, and the moment when the potential of AI tools finally clicked for her. 🎧 🖥 Listen on Spotify or Apple Podcasts, or watch on X or YouTube.

“When Your Vibe Coded App Goes Viral—And Then Goes Down” by Dan Shipper/Chain of Thought: Every CEO Dan Shipper vibe coded our new document editor, Proof, in his spare time. But when it kept crashing on launch day, he spent 24 sleepless hours watching Codex agents hunt bugs in a codebase he didn’t fully understand. Dan’s takeaway is that if the AI can build it, it can also fix it—but it might take a while. Read this for an honest post-mortem on vibe coding at production scale.

“What Comes After LinkedIn” by Eleanor Warnock: AI is automating the routine work that many white-collar professionals built careers on, and a resume full of big-name employers no longer proves you have the judgment and taste that matter. Every managing editor Eleanor Warnock writes that knowledge workers need portfolios—collections of artifacts that show how they solve problems. Read this for the emerging portfolio formats and examples worth stealing.

“I Hired an AI to Do My Chores. Now I Maintain the AI.” by Jack Cheng: Every senior editor Jack Cheng wanted his AI agent to handle the stuff nobody wants to do, like disputing bank charges, changing leaked passwords, clearing desktop clutter, etc. Instead, the agent kept breaking—and fixing it became another burden. The irony sent Jack down a rabbit hole about why we’re so bad at maintaining things, and what we lose when we stop doing it ourselves. Read this before you automate away your to-do list.

Log on

Upcoming camps

Claude Code for Absolute Beginners (April 14): Early bird registration is open for this beginner-friendly, live workshop led by Mike Taylor (head of tech consulting at Every), designed to get you from zero to a working project with Claude Code.

Learn more and sign up

Alignment

The n-of-1 problem. This week, a heartfelt story went viral about an Australian data analyst named Paul Conyngham who used ChatGPT, AlphaFold, and Grok to help design a personalized mRNA cancer vaccine for his dying dog, Rosie.

Rosie had advanced mast cell tumors that had beaten surgery, chemotherapy, and immunotherapy. Vets had given her months to live. So Conyngham, who has no background in biology, used AI to sequence her tumor DNA, identify the mutations driving the cancer, and design a vaccine blueprint targeting those specific mutations. He handed that information to scientists at the University of New South Wales, who created a vaccine that a veterinary immunologist then injected around the tumors. Within a few weeks, the largest tumor shrank by about 75 percent, and Rosie is now happily chasing rabbits again.

I love this story, and it went viral for obvious reasons: A man’s love for his dog meets a bold act of individual agency meets a glimpse into medicine’s AI-enabled future—this is Tech Twitter’s dream!

But what Conyngham built was a treatment designed for one dog’s specific tumors, tested on that one dog, validated by a sample size of one—also known as an “n-of-1” trial (a play on notation from clinical experiments, where the variable n represents the number of participants).

Conyngham’s n-of-Rosie trial may have worked, which is wonderful. But her cancer isn’t cured, and it’s not known how much the response came from the vaccine or other simultaneous treatments. The entire infrastructure of modern medicine, from randomized controlled trials to regulatory bodies, exists precisely because we learned the hard way that anecdotes aren’t evidence.

Yet we’re drifting into n-of-1 territory everywhere. Some people are injecting peptides sourced from Chinese manufacturers and self-reporting benefits that could easily be placebo, while others are going to Mexico and Colombia for untested gene therapies that they believe will prevent aging.

What happens to evidence-based medicine when personalized treatments become trivially easy to design but remain extremely difficult to validate? Ethics committees, which approve and oversee lab experiments on humans and other animals, took Conyngham months to navigate and are not built for a world where thousands of people are desperate to use their own AI-generated blueprints.

Many people will understandably look at Conyngham’s example and be emboldened to take a similar approach to their own health. I also suspect that, because of Rosie’s story, bioethicists are losing quite a bit of sleep.—Ashwin Sharma

That’s all for this week! Be sure to follow Every on X at @every and on LinkedIn.

For sponsorship opportunities, reach out to sponsorships@every.to.

Upgrade to paid

When Your Vibe Coded App Goes Viral—And Then Goes Down

Dan Shipper / Chain of Thought — 2026-03-20 10:00:00 -0400

by Dan Shipper

in Chain of Thought

Midjourney/Every illustration.

Was this newsletter forwarded to you? Sign up to get it in your inbox.

At 4 a.m. on the day after we launched our agent-native document editor, Proof, I watched yet another Codex agent try to revive our server.

Over 4,000 documents had been created since launch, but the app had been mysteriously crashing all day. This left users with crucial documents that they couldn’t access, and me with egg on my face.

I hadn’t slept for almost 24 hours, and all I could do was nervously munch trail mix as Codex investigated yet another bug buried deep in a codebase that I didn’t understand. It felt less like programming and more like being the dumbest participant at a math Olympiad. Needless to say, I was reconsidering my life choices.

Today, almost a week later, Proof is more or less stable. And I’ve learned a lot about both building and launching a purely vibe coded app. Perhaps more importantly, I’ve also learned what happens once that app goes live—and then goes down.

My current opinion is this: If you can vibe code it, you can vibe fix it. You just might not be able to fix it quickly.

Software engineering is changing rapidly as a discipline. The days of typing code into a computer manually seem to be over, and the current conversation on X is around “zero-human startups.” My experience with Proof, though, is a good reality check.

It demonstrates both what is truly possible with vibe coded apps, and where human engineers will continue to be critical now and in the future.

What’s possible at the edge

I’ve been writing about how AI is changing programming for a few years now, and my experience with Proof confirms a lot of my thoughts:

You can vibe code and launch a complex app extremely quickly

I built Proof on the side while running a company of about two dozen employees.

I first committed code to Proof’s Github repository in early January. Back then, it was a MacOS app. Two and a half weeks ago, I pivoted it to be web-only. The final form that we launched was about 10 days old, and in that time, I’d built a version that was stable enough that it had become a critical internal tool at Every, and was being used by a small cohort of enthusiastic subscribers, too.

The repository now has 1,600 commits—the vast majority from me—over 600 pull requests, and about 140,000 lines of code. To give you a point of comparison, between 2020 and 2024, the median full-time developer committed code 417 times per year.

Vibe coding is a new superpower for everyone—but people still underappreciated its importance for CEOs.

Codex is a powerhouse

I primarily used OpenAI’s Codex desktop app and GPT-5.4 to build Proof. I find GPT-5.4 to be faster, more personable than previous Codex models, and extremely smart. Where other models like Anthropic’s Opus started to fail as my codebase grew, Codex hummed along nicely.

As of this week, Codex also has subagents, which lets you hand off work to specialized agents that run in the background. Subagents dramatically accelerated my work. As I was trying to keep our server from going down, I’d have one subagent pushing the latest fixes live, another watching for new errors, and another coding up a solution to the highest-priority issue, all while an orchestrator was managing them for me.

I also used Every’s compound engineering plugin extensively as the project got further along. Its ability to make deep, well-researched plans and its comprehensiveness in reviewing code only further extend Codex’s power.

The new bottlenecks at the edge

In launching Proof, it also became obvious to me where coding models still fall short—and what new bottlenecks show up now that we don’t have to type code character by character into our computers anymore. Here’s a list:

Coding models make a mess when they try to fix issues in isolation

Coding models often prefer to fix the local issue instead of stepping back first to identify the root cause. Without your guidance, the codebase risks becoming a patchwork of hot fixes that cause more problems than they solve.

Coding models don’t always know best practices

Proof is built on Yjs and Hocus Pocus, two open-source libraries for building live, collaborative document editors. A week into the latest version of the app, I realized that the best practices around Yjs aren’t in GPT-5.4’s training data. This meant it wasn’t catching some obvious architectural deficiencies, and was pursuing suboptimal solutions to certain problems.

Asking the model to do significant web research before building seemed to mostly fix the issue. But this relies on the best practices being on the web, which is not always the case—especially if you’re working in enterprise software and using private company libraries.

Coding models will figure out the right answer—eventually

Even with the above problems, I do think Codex would have eventually solved most issues. But coming to the right answer requires a feedback loop.

Once Codex, or any coding model, writes a fix, it needs to review the code, test the fix locally, test the fix in a staging environment, get the code into the production app, and then monitor to see if the fix is working. On a small codebase, this loop is fast. On a large, complex one, it can take hours.

Why human engineers aren’t going away

Every bug fix is a scientific experiment. A coding model has no way to shortcut the cycle—of hypothesizing a fix, coding it up, deploying, monitoring, and, if the fix doesn’t work, forming a new hypothesis. It can only run the experiment and wait. The cost, in time and tokens, grows with the complexity of your app.

An experienced engineer using a model runs fewer experiments. They look at a broken system and, based on years of their own failed experiments, quickly narrow the possibilities—this is probably X. These hunches aren’t always right, but they are right often enough to make the difference between an immediately stable app and one that’s broken for a few days while your agent hunts down the issues.

This is the whole point of the allocation economy. The scarce resource is knowing where to direct intelligence. Expertise increasingly means being able to ask the right first questions, narrow the hypotheses, and waste fewer cycles.

And faster model progress doesn’t get rid of this. Because the frontier of expertise is moving even as new models are being trained. The best practices for using AI on complex production systems are being worked out right now, by engineers doing it, and those practices won’t show up in training data for months or years—if at all.

So yes, we are now living in a world where if you can vibe code, you can vibe fix it. But without human expertise, it might take a while.

Dan Shipper is the cofounder and CEO of Every, where he writes the Chain of Thought column and hosts the podcast AI & I. You can follow him on X at @danshipper and on LinkedIn.

To read more essays like this, subscribe to Every, and follow us on X at @every and on LinkedIn.

We also do AI training, adoption, and innovation for companies. Work with us to bring AI into your organization.

For sponsorship opportunities, reach out to sponsorships@every.to.

How to Build an AI Style Guide

Katie Parrott / Guides — 2026-03-19 09:00:00 -0400

by Katie Parrott

in Guides

Midjourney/Every illustration.

Update: On April 14, 2026, we’re running Claude Code for Absolute Beginners—a full-day, beginner-friendly course designed to get you from zero to a working project with Claude Code. The workshop is led by Every’s head of tech consulting, Mike Taylor, with Every CEO Dan Shipper. No programming experience is required.

In 1893, a printer named Horace Hart pinned a piece of paper to the wall of the Oxford University Press room where books were printed. It told his compositors—the workers who assembled metal letters by hand into the trays that would be inked and pressed against paper—how to handle the decisions that came up hundreds of times a day. When to capitalize. Where to put a comma. How to abbreviate, hyphenate, and spell.

That piece of paper grew into a 200-page manual, which people called Hart’s Rules. It’s now one of the oldest continuously maintained style guides in print.

The term “style guide” itself didn’t appear in print until the 1930s. For most of publishing history, people said “style manual” or “style sheet.” But whatever you called it, the function was the same: Make the rules explicit so the output stays consistent, no matter who—or what—is producing it.

When we hear the word “style guide,” we tend to think about rulings on the Oxford comma or how to abbreviate “United States.” But Hart was solving a machine problem. Oxford was printing 272 books a year, and dozens of compositors worked the presses. His guide made sure output stayed consistent no matter who was at the press that day.

Every time a new machine changed how writing was produced and distributed, someone wrote a style guide to bridge the gap between what the writer intended and what the technology could do. When the telegraph arrived in newsrooms and could cut off mid-transmission, reporters learned to front-load the essential information, producing the inverted pyramid—the journalistic principle that the most urgent idea in a story should come first, followed by less pressing details. Eventually, we got the AP Stylebook. When screens changed how people read, web style guides made everything shorter and scannable.

Language models have opened a new gap. They can generate writing, but they can’t generate yours. Left to their defaults, they converge on the same polished, generic, interchangeable prose—the kind that opens with “In today’s rapidly evolving landscape” and closes with a three-part summary no one asked for. And once again, the solution is a good style guide. Like Hart’s Rules or the AP Stylebook before it, an AI style guide takes what the writer knows and puts it in terms the machine can use.

Surfacing the hidden patterns

Previous style guides focused on issues that were solvable with conscious, mechanical decisions. Hart could tell his compositors to spell “colour” with a u, and the job was done. The AP could tell reporters to lead with the most important fact.

In contrast, the most distinctive markers of a writer’s style tend to be subconscious: word choice, sentence rhythms, and patterns that fall into place while you’re focused on meaning. Researchers have found that humans are roughly twice as linguistically varied as machines, sentence to sentence. The gap narrows with better models, but it doesn’t close.

The only way to make a model approximate your unique voice is to surface those hidden patterns, name them, and write them down as instructions.

I stumbled into this when I started feeding my essays to ChatGPT and asking it what it noticed. Over weeks, it surfaced patterns I’d never consciously seen, such as how I structure an argument to progress from personal experience to broader stakes, or tend to favor long, winding sentence structures (a temptation I’ve calibrated my style guide to help correct). I copied what felt true into a Google document and continued the conversation, pushing back on some observations and adding in details I knew I wanted my writing to avoid. Eventually, I uploaded it all to a Claude Project, and the writing that came back started to sound like me.

It’s surprisingly hard to describe your own voice from a blank page. You know what your writing sounds like. You can recognize it instantly. You just can’t explain it in terms a model can use—and vague descriptors like “smart” and “conversational” don’t make the AI better at approximating you.

It’s all well and good to say, “You should make a style guide,” but we also know that when you’re trying to get somewhere, it helps to have a map. So we wrote a step-by-step guide to building your own AI style guide from scratch—what it is, what goes in one, how to build one by letting AI interview you, and how the practice can grow with you and your writing process. It draws on the systems we’ve built at Every and for Spiral, our AI writing tool. By the end, you’ll have a working document you can hand to your chatbot, agent, or OpenClaw assistant and test on your next draft.

Hart’s compositors didn’t need to understand why they spelled “colour” with a u. They followed the rule. But writing an AI style guide forces you to answer a harder question: Why do I write the way I do? The document you end up with may be addressed to a machine, but the person who learns the most from it is you.

Read the guide

We also do AI training, adoption, and innovation for companies. Work with us to bring AI into your organization.

For sponsorship opportunities, reach out to sponsorships@every.to.

Editing AI Writing

Every Staff / Context Window — 2026-03-18 09:00:00 -0400

by Every Staff

in Context Window

Midjourney/Every illustration.

Was this newsletter forwarded to you? Sign up to get it in your inbox.

‘AI & I’: How Every builds a writing team in the age of AI

Today, we’re releasing a new episode of our podcast AI & I. Dan Shipper sits down with Every’s editor in chief, Kate Lee, to discuss how she views AI as an editorial leader and how she uses it daily. Kate’s career has spanned a stint as a New Yorker-featured literary agent to roles at Medium, WeWork, and Stripe.

Contrary to his “early adopter” persona, Dan classifies Kate as a “pragmatic knowledge worker,” someone open to AI, but who isn’t going to immediately change her workflow unless a tool makes her life better.

Watch on X or YouTube, or listen on Spotify or Apple Podcasts to learn what tools have convinced Kate. You can also read the transcript.

Here are the highlights:

AI adoption clicks when it solves a real pain. Kate’s AI “aha moment” was when she used an agent on the Atlas browser to handle a dreaded Notion setup for hiring. The AI gave her a first pass on candidates and handled the administrative work, which made it possible to hire for multiple roles even when she had hundreds of applicants and no human resources department.
Codifying taste into AI is the new editorial superpower. Kate built a 400-rule style guide and fed it into a Claude project so that writers and editors could check drafts with it before they reach her for a final check. Every piece arrives at Kate in better shape, freeing her to focus on whether a piece is the best it can be for Every rather than catching mechanical errors.
Small teams can now do what big teams did, but it requires a certain mindset. Every went from four to 20 people while dramatically expanding its offering. Kate emphasizes that the step change happened around late 2024 and early 2025 when more powerful models and tools like Claude Code and Cowork emerged. That growth is only possible if you’re willing to learn new workflows and learn from others, she says.

Miss an episode? Catch up on Dan’s recent conversations with LinkedIn cofounder Reid Hoffman; the team that built Claude Code, Cat Wu and Boris Cherny; Vercel cofounder Guillermo Rauch; podcaster Dwarkesh Patel; and others, and learn how they use AI to think, create, and relate.

The machine translation problem

A friend recently translated a healthcare app into French. “It was the most painful work I’ve ever done,” she told me. Instead of asking for a full translation, the app’s creator gave her a machine-translated version to correct, arguing that it would cost less money. Given how poor the writing was, it would have been far quicker—and better—to translate the app from scratch.

I’ve felt the same way when I edit writing clearly generated with AI. Just like my translator friend, I often feel as if it would be easier to write it from scratch rather than trying to save it with an edit.

Besides the tells (staccato, lists of three), AI-generated text is flimsy. Poke it just a bit—what do you really mean?—and it falls down. It is the opposite of what I call “bulletproof writing,” a style that was drilled into me as a financial reporter at the Wall Street Journal. Each word printed was scrutinized by tough editors and even tougher readers—so they had to be intentional.

This could be avoided, or at least mitigated. Our staff writer Katie Parrott recently shared her process for using AI to write, and the most striking thing was how much work she does before drafting. She fed Claude examples of her writing, had it interview her about her preferences, and produced a style guide that lives inside a dedicated project. She treats the whole thing like a bonsai garden—she prunes old examples, adds new ones, and reruns the analysis. With this upfront investment, Claude has her DNA when she sits down to write. As someone who often edits Katie, I can tell the difference. The writing feels like her. Kate, our editor in chief, has also codified Every’s style guide in a Claude project that everyone can use, as she talks about in this week’s podcast.

Reading matters, too. It teaches you what good writing is—something Katie also believes. So before you summarize an article with ChatGPT, think again. What do you miss when you skip the actual text? Study the structure, the argument. Steal it.

Writing is still hard. Don’t let AI make you think it’s easy.—Eleanor Warnock

Log on

Upcoming camps

Claude Code for Absolute Beginners (April 14): Early bird registration is open for this beginner-friendly, live workshop led by Mike Taylor, head of tech consulting at Every, designed to get you from zero to a working project with Claude Code.

Recordings you may have missed

Compound Engineering Camp: Cora general manager Kieran Klaassen showed how he builds with the compound engineering plugin, and walked through, step by step, the process of going from a single prompt to a working app in under an hour. Watch the recording or read the write-up.
OpenClaw Camp: The Every team walks through OpenClaw from the ground up, showing step-by-step setup and the team’s favorite use cases. Watch the recording or read the write-up.

Straight from Slack

A third of users who buy Sparkle’s lifetime plan are coming through ChatGPT. (Screenshot courtesy of Every.)

If you needed another reminder of the power of model context protocol (MCP), here it is. We’ve always had the product analytics platform PostHog installed, but linking it to AI like Claude through an MCP has given the team even deeper insights on which to base product decisions by asking simple questions in plain English. That data sourcing and interrogation would previously have taken hours. For example, PostHog data showed us that 33 percent of buyers of the lifetime plan of our file organization software, Sparkle—which costs $279—come through ChatGPT in the last 30 days. (The monthly plan is $15.)

The PostHog data fed through the MCP also helped the team notice that existing customers were pausing lifetime plans and then restarting them, suggesting that they were researching and considering the product.

“The PostHog MCP data flow is now effortless. It helps us to make decisions and cross-check them before we take them live,” says Sparkle general manager Yash Poojary. “You now have a top-tier product manager available to you. You just need to know which question to ask them.”

The new version of Sparkle launches on April 14 and will allow users to customize and create their own folder structure. Existing users will be upgraded automatically. The latest version is available already.

To read more essays like this, subscribe to Every, and follow us on X at @every and on LinkedIn.

We build AI tools for readers like you. Write brilliantly with Spiral. Organize files automatically with Sparkle. Deliver yourself from email with Cora. Dictate effortlessly with Monologue.

We also do AI training, adoption, and innovation for companies. Work with us to bring AI into your organization.

For sponsorship opportunities, reach out to sponsorships@every.to.

I Hired an AI to Do My Chores. Now I Maintain the AI.

Jack Cheng — 2026-03-17 10:00:00 -0400

by Jack Cheng

Midjourney/Every illustration.

AI agents promise to automate away the tediousness of modern life—the overbilled rental cars, the iCloud storage alerts, the changing of leaked passwords. Jack Cheng, Every’s senior editor, put that promise to the test. But instead of his AI agent maintaining his digital life, he ended up maintaining his AI agent. From there Jack explores Stewart Brand’s philosophy of “nested maintenance,” COBOL Cowboys, and civic technologists watching Claude Code attempt to modernize government benefits systems. Read on for an account of what it means to hand our most tedious obligations to machines—and what we only come to understand about broken systems by struggling with them ourselves.—Kate Lee

Was this newsletter forwarded to you? Sign up to get it in your inbox.

I set up OpenClaw in hopes that it would automate away the petty bureaucracy of modern life.

Maybe a Claw could keep my iCloud storage account from constantly hitting its cap, or go through my over 1,000 different online accounts and change all the passwords that were leaked by hackers onto the dark web. Maybe, I thought, it could even help my family sort out a medical bill we got from an unexpected hospital visit while traveling before we switched from our old health insurance to our current one.

It’s not the first time I’ve tried to tackle this problem. Since 2023, I’ve been hosting what I call a “Digital Mending Circle.” With a small group on Zoom, I tend to the maintenance tasks that accrue around a digital existence. Instead of darning socks or patching jeans, we update personal bios, organize photos, file expense reports, or even just catch up on email.

These activities can feel surprisingly daunting, given how trivial they are in the grander scheme. They involve re-familiarizing yourself with systems you only use occasionally (where’s that page in my WordPress admin panel again?) or facing clean-ups—the 571 items on your desktop, the gigabytes of blurry and duplicate photos across multiple apps—that will just need re-tidying months later. Maybe that’s why we so often neglect them.

Now, generalist AI tools like Claude Cowork and specialist tools like Sparkle, Every’s AI file organizer, can do many of these tasks for you—and swiftly. They’re chores that Claws, or whatever forms personal AI agents take in the future, could do for you without your ongoing input.

So I’ve been pondering this question: What does maintenance look like when you have AI running your digital life?

The maintenance of everything

“Maintenance is absolutely necessary and maintenance is optional,” says Whole Earth Catalog and Long Now Foundation founder Stewart Brand in Maintenance: Of Everything, Part One. Optional because we can put it off in the moment, necessary because putting it off for too long can lead to disaster.

We learn, in the book, of boats whose maintainability resulted in very different outcomes for three sailors competing to first circumnavigate the globe. We discover how maintenance attitudes in militaries can sway entire wars. Good and poor maintenance can both have profound consequences.

Maintenance is virtuous. But it’s also rarely seen as heroic. If it were, maybe we wouldn’t be so bad at it. Various explanations exist for why we deprioritize maintenance, ranging from cultural values (we prize new invention over care for the existing), psychology and economics (we discount what isn’t immediately gratifying), and social class (we associate many maintenance jobs with minimum-wage work done by marginalized workers).

For me, many of these attitudes are embodied in Pixar’s 2008 animated film about a solitary garbage robot, Wall-E. The cheerfulness with which Wall-E performs his Sisyphean task of collecting and compacting tiny robot-sized cubes of trash makes us care for him. But he doesn’t become a hero until he leaves behind his duties to follow his love interest across the galaxy. At the end, we learn his quest is part of a larger story of failed maintenance—of the earth and its natural systems.

“Nearly everything worth maintaining is nested,” writes Brand, “in something larger, even more worth maintaining.”

Admin nights

For over six years, journalist and author Chris Colin has been hosting in-person gatherings akin to my digital ones. At these “Admin Nights,” Colin and friends gather with their laptops to cancel streaming service subscriptions, file insurance pre-authorization forms, dispute erroneous credit card charges, and, more generally, try to pull themselves out of the morass of maddening tasks that swallow modern life.

In the process, they’ve grown more aware of the sources of that madness, like the rise of subscription models and the breakdown of unions, regulators, and community groups that once shielded us from consumer abuse. Will AI eliminate this administrative friction, or only worsen it?

Colin does not dismiss the possibility that AI could make a meaningful difference, but said the group has developed a jaundiced long view of new technology. “I came to San Francisco in 1998, and I’ve seen a lot of things come through that promised to be amazing and weren’t,” Colin said to me over the phone. “Or things that promised to be amazing and were, and then were discontinued or evolved or broke after a few years.”

Even if AI ends up being “amazing,” I also wonder if smoothing over the friction would obscure its underlying causes, like the convenience-laden hoverchairs in Wall-E that numb senses and distract the humans from their world’s ills. Maybe it’s through the act of collectively dealing with that friction that Colin and his Admin Night group start to understand—and perhaps change—the broader systems to which most of us belong.

The COBOL cowboys

Those broader systems are facing a maintenance crisis of their own. Consider COBOL, a programming language designed in 1959, that handles transactions in 43 percent of all banking systems worldwide. It runs primarily on IBM mainframe computers, and still underpins much of the infrastructure behind public benefits like unemployment and food assistance that millions depend on.

Institutions haven’t modernized their COBOL systems because the process is a pricey, years-long affair. Meanwhile, the number of people in the world who have the expertise to maintain these systems shrinks by the month. When the situation is dire, organizations often call on the COBOL Cowboys, run by 82-year-old cofounder and CEO Bill Hinshaw and staffed with some 600 COBOL engineers, many of whom got their start in the 1960s and ’70s programming those original mainframes. Heroes, for sure, but maybe only because they come in when these systems break.

So when Anthropic announced last month that Claude Code can now automate the most time- and cost-intensive aspects of the COBOL modernization process, IBM’s stock fell by 13 percent, its single largest drop in 26 years.

Not great for IBM’s business, but potentially transformative for people stuck on the other side of these systems. Adam Selzer, co-founder and senior director of Detroit non-profit Civilla, which helps local and state governments redesign how they administer benefits to their constituents, says that legacy technology is a significant barrier to change within these institutions.

“[It] could be one of the most significant developments in civic technology in this decade,” says Selzer of the Anthropic news. “Or could lead to it.”

What’s Claw got to do with it

I personally haven’t gotten as far as overhauling the economy’s financial infrastructure. I’m still trying to contact my bank about an overbilled rental car and make my “iCloud Storage Full” alert go away.

So at a recent Digital Mending Circle, on the promise of a more frictionless life, I set up OpenClaw. Or started to set up is more accurate. It took several hours beyond my allotted life maintenance session for me to get my Claw, Pip, working. Once Pip was up and running, I had it send my partner and me morning briefings of our day’s schedule and childcare duties, and prepare budget summaries that we could go over in our weekly household meeting—augmenting a system we already had in place to help us stay afloat.

But sometimes Pip would format the budget summary in a weird way or forget how to access our transactions. Sometimes the morning briefings wouldn’t come—a result of my Claude Code account hitting its monthly limit or Pip incorrectly updating a setting in its own configuration file. At one point, two weeks’ worth of Pip’s memories simply vanished. In each of these cases, I would have to then go in and figure out why Pip broke, and how to fix it.

Instead of Pip helping maintain my digital life, I was maintaining Pip.

But in the process, I’ve become intimately familiar with the OpenClaw configuration file. I’ve come to better understand the cascade of markdown instructions that Pip reads for context, the art of managing Pip’s short- and long-term memory, and the way it uses skills and tools to do its work.

I’ve also come to learn, true to Brand’s words and Colin’s Admin Night group, the wider context around Pip: the ecology of frontier models, AI agents, new and existing software, and the people trying to use and build these tools. And once I could see this more clearly, I could start to imagine these systems working and interacting in a different way.

Then again, I still have to look into that rental car bill.

Jack Cheng is a senior editor at Every. He is a creative generalist and the author of critically acclaimed fiction for young readers. He writes an occasional newsletter called Sunday.

To read more essays like this, subscribe to Every, and follow us on X at @every and on LinkedIn.

We also do AI training, adoption, and innovation for companies. Work with us to bring AI into your organization.

For sponsorship opportunities, reach out to sponsorships@every.to.

What Comes After LinkedIn

Eleanor Warnock — 2026-03-16 08:00:00 -0400

by Eleanor Warnock

Midjourney/Every illustration.

Was this newsletter forwarded to you? Sign up to get it in your inbox.

Whether or not you believe that flawed Anthropic graph or Jack Dorsey’s stated reasons for laying off 40 percent of his staff at Block, the message is the same: AI is doing more of the rote click-work that sustained the laptop class. The knowledge workers who thrive will be the ones who can add something extra—judgment, taste, and expertise.

The problem is that most of us have no good way to prove those things, and we’re not very good at it. For decades, brand names did the heavy lifting. Previous experience at McKinsey or Goldman Sachs were buzzwords that signaled to a hiring manager: This person can build a model and make slides, and someone selective chose them.

But brand names don’t work as a heuristic for expertise anymore. Was that Google alumnus a pixel pusher or a genuine decision maker? The soft skills that matter most in an AI-augmented workplace are invisible on a traditional resume.

One answer is portfolios. Not a portfolio in the sense of a portfolio career—a word that many knowledge workers have embraced as they move away from relying on one employer and pick up advisory roles and solopreneurship. I mean a portfolio as a body of work that proves your value beyond your work history. A collection of artifacts that show how you think, just like creative professionals showcase their design or artistic work.

As a journalist and editor, I write a newsletter about people using writing to build things, and I post on LinkedIn, all of which I consider to be part of my portfolio that helps people evaluate me beyond the names of my previous employers. Whether it is writing on platforms like LinkedIn or building interactive tools on their personal websites, knowledge workers need something similar. AI can make this process fun and creative, and help people experience your expertise without having to meet you.

Wasn’t personal branding supposed to fix this?

Even before anyone was worried about ChatGPT taking their job, we’d been hearing the same advice for years: Build your personal brand. Post on LinkedIn more, put yourself out there—become a thought leader. In theory, a strong personal brand should give you an advantage over someone with the same credentials on their resume but no public presence.

In practice, almost nobody does this well. LinkedIn has more than 1 billion registered users, but only a tiny fraction of those share content. Much of what gets shared is self-congratulatory announcements, not proof of expertise. And more and more of what is shared is obviously AI-written. Readers’ eyes gloss over the drivel as they move to the next rocketship emoji on their timeline.

The people who do want to share real insights—writing-first practitioners, as I’ve been calling them—are fed up with the platform. You can’t format text in posts. The algorithm surfaces two-week-old content like it’s breaking news, and anything older gets buried. That’s why wealth manager Jan Voss moved to a newsletter to share his most important thoughts. He wanted a place where, as he told me earlier this year, “posts have staying power.”

The platform’s functionality is only part of the problem. Thinking more deeply, LinkedIn rewards a specific kind of extroversion. Many people feel uncomfortable with building their brand there because they feel pressure to have a “hot take” or, more simply, to brag.

The age of the portfolio is coming

We’re entering a period where knowledge workers build portfolios the way creative professionals have for years—not on Dribbble or Behance, but through a growing ecosystem of formats that let you show how you think.

Substack and similar platforms are making a play here, letting individuals share deep thinking that endures and build a following around their expertise. Substack even wants to let users express themselves beyond writing, as last week’s launch of a video recording and publishing feature shows. But if the hurdle to writing frequently on LinkedIn is high, the hurdle to maintaining a regular newsletter or recording a regular podcast is enormous. You need to have information or opinions to share with your readers each week, even if you are simply curating links.

One Substack writer who committed himself to one post a week for a year wrote 62,773 words, or the equivalent of a short novel. Most people don’t have the stamina or the discipline to achieve that level of prolificness. Hence, Substack as a reputation-building tool will remain limited to the small minority of people willing to write consistently in public.

Other formats are emerging. Platforms like Andreessen Horowitz-backed Maven offer cohort-based courses where you can teach what you know to a room full of people who’ve paid to learn it, and given that teachers must apply, being featured on the platform carries credibility. Intro, also backed by Andreessen, does something similar with one-on-one calls.

And then there’s a newer category: knowledge workers building interactive tools on their personal websites that give potential employers or clients a taste of what they can do.

Strategist Loris Colantuono has a section called “Resources” on his website with guides about storytelling and manifesto writing—a link at the end prompts visitors to set up a call. Julius Bachmann, a Berlin-based executive coach, did something similar with his annual review document at the end of last year. He released the guide as a PDF and a custom GPT so you can either fill it out on paper or by chatting with AI.

Both examples are interactive proof of how these individuals think. Even if you never pay either of them, you’ve seen what they have to offer. It’s also scalable. A thousand people can experience Loris’s strategic thinking or Julius’s coaching style without them having to take an equal number of coffee meetings.

Building these kinds of tools and portfolios can be less stressful for people who are uncomfortable with LinkedIn’s self-promotional tone. You don’t need to perform or humble-brag. You can create something on your own terms that looks and feels like you.

Building your portfolio with AI

The same technology that’s threatening knowledge work is also lowering the barrier to building these kinds of portfolios. A year ago, creating an interactive tool on your personal site required hiring a developer. Now you can spin one up in an afternoon.

Katie Parrott, a staff writer at Every and AI editorial lead, redid her personal website in early 2025 after a colleague encouraged her to try out Loveable. Though the process was clunky, “it was the first time the site felt like mine,” she says.

Katie used Lovable to redo her personal website in early 2025. (Screenshots courtesy of Katie Parrott.)

She rebuilt the site again in February 2026 as a way to test the capabilities of her new personal AI agent, Margot. “I pointed her at a folder containing my entire ChatGPT and Claude conversation histories and said: ‘Based on everything you know about me, come up with an updated concept for my website.’” Margot came up with a design that highlighted how Katie made things with words—typing animations, blinking cursors in unexpected places.

Her website brings together all her projects (Every writing, Substack, an AI writing tool she’s building) and explains clearly who she is as a professional, with panache—“Writer. Builder. Chronic overthinker.” Her favorite feature, one she came up with, also provides a glimpse at what’s important to her: a changelog. It’s a running, curated inventory of things she has published, built, and learned that don’t necessarily warrant their own post but that she wants to capture. “For a long time, a personal website was a static credential. Now, it’s more like a living document of momentum,” she says.

Katie publishes lessons, new articles, and updates to her tools on a log on her personal website.

There’s at least one obstacle, though: A lot of the best expertise lives behind NDAs and confidentiality walls. If your most impressive work is a strategy you built for a client you can’t name, using data you can’t share, you have no proof of work when the world asks. Perhaps this shift toward portfolio-building will create pressure on NDAs to allow at least anonymized sharing of real work.

The trust problem

You are probably thinking: Portfolios are great, but how do you prove the judgment was yours? Or prove that you didn’t ask Andrey Galko, who manages Every’s website, to create something impressive for you?

To this end, I think that a rise in portfolios will be accompanied by more paid work trials. Not the unpaid case study interviews (do a presentation about our strategy, the output of which we will promptly steal once we tell you you don’t have the job) that have rightly been criticized, but short, compensated engagements where you work alongside a team for a week or two before anyone commits. The trial proves those skills in your portfolio are real.

I used to put a hold in my calendar to update my resume every few months. Now, I’ve put a hold to create a personal website and portfolio.

Eleanor Warnock is the managing editor at Every. She has been a business journalist and editor at the Wall Street Journal and the Financial Times-backed Sifted, and is an advisor to Bek Ventures. Follow her on LinkedIn and Substack.

To read more essays like this, subscribe to Every, and follow us on X at @every and on LinkedIn.

We do AI training, adoption, and innovation for companies. Work with us to bring AI into your organization. Discover Every’s upcoming workshops and camps, and access recordings from past events.

For sponsorship opportunities, reach out to sponsorships@every.to.

The Never-done Machine

Every Staff / Context Window — 2026-03-13 15:00:00 -0400

by Every Staff

in Context Window

Midjourney/Every illustration.

Hello, and happy Sunday! Was this newsletter forwarded to you? Sign up to get it in your inbox.

Knowledge base

“Introducing Proof” by Dan Shipper/On Every: We released a new product: Proof is a free, open-source document editor built for agents and humans to collaborate, with live editing, comments, change tracking, and simple visual cues that show who wrote what. At Every, we use it for everything from product plans to daily to-do lists. Read this to see how it works and try it yourself with a ready-made prompt for your agent of choice.

“AI Was Supposed to Free My Time. It Consumed It.” by Katie Parrott/Working Overtime: Every staff writer Katie Parrott sat down at lunch to work on a project with her new AI assistant and found herself prompting away until 1 a.m., a pattern that’s become all too common for her and, as she discovered, plenty of others. Instead of reducing work, AI makes people want to do more of it, through task expansion, blurred boundaries, and a slot-machine dopamine loop. Read this for the psychology behind AI compulsion and tactics to help break the cycle.

“The Science of Why AI Still Can’t Write Like You” by Marcus Moretti: AI can demonstrate Ph.D.-level knowledge, but its writing remains stubbornly detectable, writes Marcus Moretti, the new general manager of our writing app, Spiral. New research reveals why: The most distinctive fingerprints of your prose come from subconscious choices—articles, pronouns, and function words that text analysis commonly filters out. Humans are also twice as varied in their writing as machines. Read this for what the science of style means for the future of AI writing tools.

“Compound Engineering Camp: Every Step, From Scratch” by Katie Parrott/Source Code: At Every’s first Compound Engineering Camp, Cora general manager Kieran Klaassen went from a one-line prompt to a working app in under an hour. He walked subscribers through every phase of the loop—brainstorm, plan, work, review, compound—showing how each step’s output feeds the next and why he spends 70 percent of his energy on planning. Read this for the full live walkthrough and advice on the best models to use for each step.

“How Main Street Companies Are Using AI” by Sam Gerstenzang/Thesis: Former Stripe product leader Sam Gerstenzang runs a funeral home and a medical spa platform—not exactly Y Combinator darlings. But as software gets easier to build, Sam (who writes his own newsletter) argues these operationally complex, real-world businesses are where AI can have the greatest impact. One of his most surprising findings is that when his team replaced human receptionists with AI, customers left faster—even though the error rate was identical. Read this for a grounded playbook on bringing AI to Main Street.

Log on

From the field. On Monday, March 16 at 3 p.m. ET, join Every’s head of tech consulting and regular columnist Mike Taylor and editor in chief Kate Lee for a livestream about everything he’s learned about teaching Claude Code to beginner-level students.

Get notified for the livestream

From Every Studio

Cora opens up to your AI agents

Cora now offers API tokens so you can connect your AI agents—Claude Code, OpenClaw, and others—directly to your inbox. Kieran built this for the growing number of users who want their agents to pull context from email without switching tools. From the new Agents page, you can set up a token, point your agent at Cora, and let it answer questions about your inbox on your behalf.

Spiral gets smarter about learning your voice

Spiral launched two new ways to build a personal style guide—no copy-pasting required. Connect your X/Twitter account and Spiral pulls up to 1,000 of your recent tweets, weighted by engagement, to tune your work to what resonates with your audience. Or paste in any URL—a page, a full site, an RSS feed, or a sitemap—and Spiral automatically builds a style from up to 20 recent posts. Marcus also gave the editor a full polish: Spiral remembers your last-used style, autoscrolls as new text comes in, and handles attachments more intelligently. Try it out at writewithspiral.com.

Collaborative filtering

Code-free. Earlier this month, Dan Shipper was the guest on New Economies’s Big Ideas podcast, where he talked about how he vibe coded Proof between meetings, how always-on agents are shaping the way our team works at Every, and why media distribution is the key competitive advantage now that software has become free to build. Watch or listen.

Alignment

Digital twins. When I was 16, I spent a summer working with my dad in a welding factory. Apart from the acrid smoke, the swearing in Polish and Russian, and the clanging that rang in my ears for hours after I left, what sticks with me most is the maze of machines, fit together like Tetris blocks on the factory floor.

I worked on a press that stamped steel into shapes that would later become supermarket shelves. The press sat near the factory door, open to the bitterly cold UK mornings. I thought, why the hell is this so close to the outside? But I knew it was there for a reason. The position and orientation of every machine on that floor had been tested and retested over years to arrive at the one arrangement that maximized output.

The efficiency of that layout was the result of hard-won experience, but today’s factory makers might use a digital twin to arrive at something like it on day one. With a digital twin, you could precisely model the factory and run countless simulations to predict what would work best. The pharmaceutical giant Eli Lilly built a digital twin of a factory to make GLP-1s—drugs like Zepbound and Mounjaro, which account for more than half of its revenue—and produced more product than they could have without AI, enough that it showed up in the company’s latest earnings report. If you can model the complexity of a pharmaceutical production line, what else can you model?

The concept works for power grids and flight networks—anywhere the variables are complex and real-world experimentation is expensive. But the application that interests me most is the human body. We already generate enormous amounts of data through blood panel results and wearables that track heart rate, sleep, and glucose. What doesn’t exist yet is the simulation layer that models your biology closely enough to test interventions before you try them.

But it’s coming, and when it arrives, your body will be what my dad’s factory floor was: a perfectly fitted Tetris puzzle, optimized for your healthiest self.—Ashwin Sharma

That’s all for this week! Be sure to follow Every on X at @every and on LinkedIn.

We also do AI training, adoption, and innovation for companies. Work with us to bring AI into your organization.

Discover Every’s upcoming workshops and camps, and access recordings from past events.

For sponsorship opportunities, reach out to sponsorships@every.to.

Upgrade to paid

Compound Engineering Camp: Every Step, From Scratch

Katie Parrott / Source Code — 2026-03-13 06:00:00 -0400

by Katie Parrott

in Source Code

Midjourney/Every illustration.

TL;DR: Cora general manager Kieran Klaassen has written prolifically about compound engineering, his philosophy of software engineering for the AI age. In this piece, based on a camp he gave for paid subscribers a few weeks ago, we get an inside look at how exactly Kieran builds with the compound engineering plugin for the first time. He walks through, step by step, the process of going from a single prompt to a working app in under an hour. If you’ve been curious about how to build with compound engineering, this is the piece to read.—Kate Lee

Was this newsletter forwarded to you? Sign up to get it in your inbox.

This time last year, any time Kieran Klaassen opened a new session in Claude Code, he started from scratch. The lessons from his past code reviews, the style preferences he’d painstakingly explained, and the bugs he’d already flagged—Kieran remembered them all, but from the machine’s perspective, it was like it had never happened.

He’d been building Cora, Every’s AI email assistant, and getting tired of copy-pasting the same prompts, correcting the same overengineered tests, and flagging the same bugs. “A human would remember,” Kieran said. “The AI wouldn’t.”

So he decided to create a system that would remember—one that plans before it codes, reviews outputs to enforce his taste, and stores every lesson so the AI applies it next time. The result is what we now know as compound engineering, a signature approach to coding with AI where every bug, fix, and code review makes the system permanently smarter. The official compound engineering plugin has more than 10,000 GitHub stars and is used by a growing community of builders, including engineers at Google and Amazon, who say it changed how they think about software.

At our first Compound Engineering Camp, Kieran walked subscribers through the full loop live, building an app from a one-line prompt to a working product in under an hour. Below is the workflow as Kieran demo-ed it, plus what it means for how software gets built from here.

Key takeaways

Brainstorm before you plan. The plugin has a brainstorm step that interviews you collaboratively and fills the gap between your vague idea and a detailed spec.
Planning should run without you. Once the requirements of the project are clear, the plugin has a plan step that researches your codebase, checks for existing patterns, surfaces past learnings, and produces an implementation plan with zero additional input needed.
Use different models for different steps. Kieran uses faster models—such as Claude Haiku 4.5 or Gemini 2.5 Flash—for brainstorming, Opus for planning, Codex for implementation, and sometimes Gemini for code review.
Compound when the context is fresh. The plugin’s compounding step stores lessons as artifacts that future agents can discover, the core of compound engineering. Run it right after something breaks or works—before the AI compacts your conversation and you lose the specifics of what you were talking about.

The compound engineering loop

A founder who does everything themselves hits a ceiling, Kieran says. The ones who scale are the ones who build systems—hiring, documenting, and training—so the work happens without them in the room.

Compound engineering applies that logic to coding with AI. “You remove yourself from as many places as you can,” Kieran said. “That forces you to extract things, automate things. And that’s where the compounding happens.”

The core idea is a four-step loop: Plan, work, review, compound. Each step has a specific input and a specific output. The output of one step becomes the input of the next.

Plan takes a problem and produces a detailed implementation plan—a markdown file (a simple text document with formatting) containing data models, file references, architectural decisions, and sources. The plan is specific enough that an AI agent or a human engineer could pick it up and execute it without asking questions.

Work takes that plan and produces a pull request (a proposed set of code changes ready for review). The code gets written, tests get generated, and documentation gets updated.

Review takes the pull request and produces findings—comments, suggestions, and flagged issues stored as to-do items in the file system. Different AI models can review the same code and surface different problems.

Compound captures whatever the system learned during planning, working, or reviewing—a new coding preference, a bug pattern to avoid, or an architectural decision worth preserving—and stores it in files that future sessions can reference. This is what makes the loop self-improving.

There’s also an optional upstream step—brainstorming—for when you have a vague idea rather than a clear requirement. “Brainstorming is when the idea isn’t super detailed yet,” Kieran said. “If you have a very clear requirement—like adding a new authentication provider—you skip brainstorming and go straight to planning. Planning is about the details and not making mistakes.”

From a blank prompt to a working app

To show how these steps connect in practice, Kieran built an app live during the session. He wanted a system that would let users vote on new features to add to Cora—a clone of FeatureBase, a platform where users submit and upvote product ideas. He started with a single prompt:

“Build a FeatureBase clone. We’ll use it as an internal feature request voting tool. Users can submit feature ideas, upvote other ideas, admins can update status. Simple Rails app with Tailwind.”

The brainstorm command turned this into a conversation. It asked questions until the requirements were clear: How should users log in? One vote per user, or multiple? How should ideas be sorted? When Kieran wanted a specific visual direction, he steered it: “I want a Swiss design vibe with square, black, white, Helvetica stuff.” Within minutes, he had a markdown document with a sketch of data models sketched, a list of pages, and an outline of a design direction.

The brainstorm step generated wireframes and page structure from the initial prompt. (All images courtesy of Kieran Klaassen.)

That document became the input for planning.

Planning is 70 percent of the job

Kieran uses Claude Code with the Opus model for the planning phase. “Opus is great for creative work, filling in the blanks, and planning,” he said. He passed the brainstorm document to the plan command and let it run.

The planner does several things that a raw AI prompt wouldn’t. It researches the existing codebase so it doesn’t reinvent what’s already there. It hunts for edge cases—steps a user might take that the brainstorm didn’t anticipate. It checks for existing design patterns that should be followed. And it pulls from compound artifacts (stored lessons from previous work) to avoid pitfalls.

The output is a markdown file with an implementation plan including phases of development, file references for where to find key details, and a detailed analysis of how the changes will impact the system as a whole. It covers both the big-picture system impact and the granular details, and it logs the source of every decision so the agent—or a human engineer—can trace why a choice was made. There’s also a “deepen” option that launches up to 80 sub-agents (specialized AI workers that each research a different aspect of the problem) for deeper research when the stakes are higher.

The planning step generated a structured implementation document outlining requirements, architecture, and workflow.

This step is where Kieran spends most of his mental energy. He estimates the split at 70 percent thinking, 30 percent doing. “Most engineers think that writing code is the work,” he said. “It’s thinking about what to even do.”

Work and review in parallel

With the plan locked, Kieran moved to implementation. He switched to faster models for this step, saving the more sophisticated ones for planning and review. The work command takes the plan and implements it: writing code, creating tests, and building the feature. If the plan has gaps, the work step flags them rather than guessing.

Within an hour of his original prompt, Kieran had a working app running locally: a clean, minimalist feature-voting tool that obeyed his design guidance.

Kieran created a working internal feature-voting tool live.

Rather than manually testing it, he ran the compound engineering plugin’s test-browser command, which uses an AI-powered browser agent to click through the interface, verify that features work, and report back.

Simultaneously, he ran the review command, using Gemini to get a separate perspective on the code. “Sometimes I let different models review the code because I get different things back,” he said. “One model finds something, and then I give it to the other model and say, ‘Is this actually reproducible and correct?’” The review step produces findings stored as to-do files for triage. He recently ran 25 agents in parallel using tmux (a terminal tool that keeps sessions alive in the background), planning, working, and reviewing across different features simultaneously.

Compound: Building a memory

The fourth step is what separates compound engineering from a sophisticated prompting habit. When you run /compound, it gathers context from the current session—whatever you learned, whatever went wrong, whatever preference you expressed—and stores it as a discoverable artifact with metadata.

These artifacts don’t bloat your context window, which is the amount of information an AI model can hold in memory during a conversation. They’re stored as files and retrieved by sub-agents only when relevant, which means you can accumulate hundreds of lessons without hitting the limits that come with stuffing everything into a single prompt.

“It’s very important to run this whenever the context is fresh,” Kieran said. “You don’t want to run it after compaction,” in other words, after the AI has started forgetting earlier parts of the conversation.

Currently, the compound step is triggered manually. Kieran is experimenting with automation, drawing inspiration from OpenAI’s harness engineering approach, which tracks recurring issues and promotes them to permanent rules after they appear multiple times. But he’s cautious about auto-generating artifacts. “It’s really easy to automatically run something and generate a lot of mess,” he said. “I still want to be in control.”

The plugin is for now, the philosophy is forever

Kieran is open about the fact that the compound engineering plugin has an expiration date. The AI labs are already incorporating ideas compound engineering pioneered—plan mode, for instance, didn’t exist in most tools when he first built his version. “Ideally, we delete the whole thing someday because it’s all built in,” he said. “The point was always the philosophy, not the plugin.”

The engineer’s job is shifting from writing code to designing systems that write code and remember what they learned. Every runs five products with single-person engineering teams thanks to that shift. The people who start compounding now will be the ones who are hard to catch later.

Install the compound engineering plugin and run /brainstorm on whatever you’re working on. Read the compound engineering guide for the full framework, and check out our upcoming courses if you want to go deeper.

We also do AI training, adoption, and innovation for companies. Work with us to bring AI into your organization.

Discover Every’s upcoming workshops and camps, and access recordings from past events.

For sponsorship opportunities, reach out to sponsorships@every.to.

The Science of Why AI Still Can't Write Like You

Marcus Moretti — 2026-03-12 06:00:00 -0400

by Marcus Moretti

Midjourney/Every illustration.

TL;DR: Why does AI writing still sound like AI writing, even as the models get smarter? In his first piece since joining Every as Spiral’s general manager, Marcus Moretti explains why the answer is more complicated than you’d think. The most reliable fingerprints of your personal style come from the words you write subconsciously: articles, pronouns, and function words that emerge in a distinctive pattern as you focus on the meaning of a sentence. His piece explores what new research in machine learning and stylometry—the study of style—means for the future of writing tools like Spiral. If you want to go deeper, Spiral has several updates, including creating a writing style from your website or X account (even taking post engagement into account) and a cleaner, faster editor.—Kate Lee

Was this newsletter forwarded to you? Sign up to get it in your inbox.

OpenAI models demonstrate Ph.D.-level knowledge across physics, biology, and chemistry. Anthropic staff have claimed its Opus 4.5 model “largely solved coding.”

Yet AI writing remains stubbornly detectable: “It’s not an idea. It’s a breakthrough.” “Delve.” Lists of threes with no “and.”

If you’re a regular Every reader, you may already know why this is. LLMs are trained on an unfathomable amount of words and learn generally how to speak. Post-training, which refines a model after initial training on large datasets, makes the models friendlier and safer, so they end up speaking in a kind of generic politeness. Ted Chiang’s description from a few years ago remains apt: “ChatGPT is a blurry JPEG of the web”—a tool that approximates human insight without ever landing on the mark.

I’m interested in the relationship between LLMs and writing style because I’m the general manager of Spiral, Every’s AI co-writer. Writing sessions in Spiral begin as a chat: You describe what you intend to write, and Spiral helps you hone your message and gather relevant research. Then it produces one or more drafts, offering several approaches for your piece.

Our aim is for Spiral’s written output to reflect your personal writing style, not the generic politeness of the foundational model. To this end, I’ve been reading papers on natural language processing, linguistic forensics, and stylometry—the study of writing styles. It wasn’t until I started working on Spiral that I became aware of the century-plus history of stylometry, or of the fastidiousness with which researchers have catalogued the elements of style. In recent years, researchers in these fields have flocked to LLMs, finding new ways to expand our understanding of human writing. Here are some findings that I found interesting and even counterintuitive, and that provide a hint as to where AI writing might be headed.

Subconscious decisions define writing styles

Stylometry has had a few moments of glory. In the 1800s, stylometrists gave sold-out lectures about whether William Shakespeare wrote those plays. In the 1960s, two stylometrists isolated Alexander Hamilton’s contributions to The Federalist Papers based largely on the presence of the word “upon.”

In the 2020s, LLMs have introduced new ways of studying style. Last year, two Cornell University researchers systematically manipulated text snippets to see how it affected LLMs’ ability to guess their authors. They removed an attribute of the text one at a time—such as proper nouns or capitalization—and measured the effect on attribution accuracy.

They found that removing the more functional features of the text caused the models to misattribute authorship more often, proving that those features are most helpful for attribution. In particular, removing “stop words” made it a lot harder to guess who wrote something. In natural language processing, stop words are common, functional words like articles (“a,” “the”) or pronouns (“I,” “she”). These words are often filtered out of text analysis because they don’t convey much meaning, but it turns out that they appear in patterns that can help identify who wrote something. This is why Hamilton’s use of “upon” tipped off those researchers to his Federalist contributions.

Things like stop words and word order turn out to be some of the most distinctive markers of someone’s writing style. These purely functional aspects of writing mostly reflect subconscious decisions. When we write, we focus on choosing meaningful words, and our subconscious tends to fill in the rest. But the way our subconscious contributes to our sentences is to be distinctive.

Most people are wildly inconsistent writers

“How Well Do LLMs Imitate Human Writing Style?” asked another paper co-authored last year by a researcher at Bucknell University. The authors used various methods to get an LLM to copy someone’s style, finding that just a few writing samples increased style fidelity over the base output by about 23 times. A smattering of examples went a long way.

They then tested whether AI text could be identified as such even if it accurately reproduced someone’s style. They discovered it can be. Natural language processing has something called a “perplexity” score, and the greater the linguistic variance, or the diversity and unpredictability of word choice and sentence structure across a writing sample, the higher its perplexity. The researchers found that, on average, humans are twice as varied in their writing as machines.

Writing styles change dramatically over time

An LLM is just a string of numbers. When you interact with ChatGPT or Claude, you’re “talking” to a static set of digital files. LLMs are point-in-time snapshots of human language, which is why they need to search the web for information after their training cutoff date.

Language itself, however, rapidly evolves. In the book Algospeak, the self-styled “etymology nerd” and linguist Adam Aleksic argues that our vocabulary is evolving faster than ever, due to social media and hyperconnectivity. This poses a problem for LLMs. What good is a model if its training run ended before we started saying “skibidi”?

In October, Sushil Khairnar, a graduate student at Virginia Tech, tried to quantify models’ “temporal drift.” He found that GPT-2 and GPT-3’s manner of speaking lagged behind the general lexicon by about 15 percent a year after its release and 28 percent after two years.

Ironically, LLMs themselves are altering language, including in academic research. Post-ChatGPT papers include significantly more AI-ese: words like “underscore,” “highlight,” and “showcase.” “Delve” is the biggest culprit, with usage in papers skyrocketing by more than 2000 percent between 2022 and 2024.

What’s next

As foundation models and methods of style transfer improve, computers will get better at mimicking individual writing styles. It’s an open question of how close they can get. The better we understand the science of style, the more we can bridge the gap between model output and manual output.

The research guides Spiral’s roadmap. As an example, we recently updated how Spiral generates a style guide from a user’s writing samples, which then stylizes Spiral’s drafts. We previously generated hundreds of descriptions of the user’s writing, but now we focus on the key textual identifiers and quintessential phrases from the source material. And we’re building connections to writing sources—blogs, newsletters, and social media—so Spiral can keep up with how your personal style evolves over time.

For any LLM-generated writing, though, there will always be some gap—after all, a person didn’t write it. That fact may be harder and harder to detect in the output, but it’s always worth considering what your reader would think upon learning that your piece was AI-assisted or -generated. This piece, for example, was written the old-fashioned way, despite the em-dashes, which Every’s style guide allows. Unless you want to analyze the stop words, you’ll have to take my word for it.

Spiral has been busy the past few weeks. New to the app: connect your Twitter account and get an engagement-weighted style guide, allowing Spiral to draft bespoke tweets for your audience. Link your website or RSS feed to teach Spiral your style via bulk post import. Workspaces now make it easy for you to share styles across your team, so you can write in one unified voice.

Marcus Moretti is the general manager of Spiral (@tryspiral).

To read more essays like this, subscribe to Every, and follow us on X at @every and on LinkedIn.

For sponsorship opportunities, reach out to sponsorships@every.to.

Introducing Proof

Dan Shipper / On Every — 2026-03-11 07:30:00 -0400

by Dan Shipper

in On Every

Midjourney/Every illustration.

Was this newsletter forwarded to you? Sign up to get it in your inbox.

If you use AI seriously to think or work, a surprising share of your documents are probably already written by agents.

Most of us at Every are using Codex to generate plan documents, Claude Cowork to write research reports, and OpenClaws to create strategy memos. But the current process for collaborating and iterating on agent-generated writing is weirdly primitive. It mostly takes place in Markdown files on your laptop, which makes it reminiscent of document editing in 1999.

Because these documents are stuck on your laptop, it’s hard for other agents to help iterate on them—and they’re hard to show to your team, too.

That’s why today we’re officially launching Proof, an online document editor built for agents and humans to collaborate. Fast, free, and no login required. (And it’s open source!)

Try Proof

What is Proof?

Most word processors still assume a human is doing the writing and AI is helping at the margins for brainstorming, making rewrite suggestions, or producing a first draft. Proof flips that around. It’s a document editor built for the kinds of documents agents are increasingly writing: bug reports, product requirement documents, implementation plans, research briefs, copy audits, strategy documents, memos, and proposals.

It supports live edits with multiple collaborators, allows you to see and leave comments, and lets you track changes.

But it also has what you would expect of a document editor built for agents. Proof:

Is agent-native: Anything you can do in Proof, your agent can do just as easily.
Tracks provenance: A colored rail on the left side of every document tracks who wrote what. Green means human, purple means AI.
Is login-free and open-source: We want Proof to be your agent’s favorite document editor.

You can try it yourself right now. Give the prompt below to your agent of choice and have it generate a Proof document for you.

Try Proof with your agent

Send your agent this link:

Try Proof with your agent

It will ask them to write a Proof document describing what they’ve learned about how to work with you best. You’ll get a Proof link with their insider’s perspective on your work style.

How we use Proof at Every

We’ve been using Proof internally for over a month and, a couple of weeks ago, opened it up to Every subscribers as an experiment. Internally, we’ve found that plans, like the ones generated when Claude Code or Codex maps out a new feature for a product, are the most common type of Proof document, along with strategy documents.

Austin Tedesco, Every’s head of growth, whose job involves writing a lot of campaign strategy documents, might go back and forth on the drafts with one or more agents. When the document feels ready, he’ll send it to the rest of the team on Slack for feedback.

Austin also uses Proof for his own writing. He sends notes for his food newsletter throughout the week to his Claw, which organizes them in a Proof document and generates an outline, so that when he sits down to write, everything is already in front of him.

I use Proof documents as my daily to-do lists. I pin the page for that day in my Slack account, and anytime a new task pops up for me to do, I’ll tell my Claw, R2-C2, to add it to the document in the appropriate section. At the end of the day, I can see which tasks I’ve completed and which I haven’t, and I can tell my agent to either push my incomplete tasks to the next day, help figure out how to complete them, or just go ahead and do them for me.

Our editorial team is also exploring how to integrate Proof into their AI workflow. The initial outline and draft of this post started in Proof.

What’s available at launch

Today, Proof is available to everyone, for free, even if you don’t have an Every account. Those who do sign in with an Every account have access to a continually updating library with all of the Proof documents they’ve created.

And it’s the first one of our products to be open-source, so that anyone can inspect it, build on it, and help shape where it goes next.

Try Proof

Most of the writing that will happen in the next decade will be generated by AI agents. Now they have a document editor built for them.

Watch me talk about Proof with Every’s Brandon Gell, Kieran Klaassen, and Austin Tedesco on X or YouTube, or listen on Spotify or Apple Podcasts on the latest episode of AI & I. You can also read the full transcript.

Try Proof

For sponsorship opportunities, reach out to sponsorships@every.to.

Help us scale the only subscription you need to stay at the edge of AI. Explore open roles at Every.

How Main Street Companies Are Using AI

Sam Gerstenzang / Thesis — 2026-03-10 04:00:00 -0400

by Sam Gerstenzang

in Thesis

Sam Gerstenzang. Every illustration.

The AI hype cycle has mostly rewarded software companies, but Sam Gerstenzang is betting on the opposite: operationally complex, real-world service businesses—funeral homes, a medical spa platform. Fresh off an appearance on our podcast AI & I, the Boulton and Watt partner and former Stripe product leader shares four hard-won lessons for injecting AI into Main Street businesses, including why humans remain stubbornly hard to replace. If you enjoy the piece, watch his episode on X or YouTube, or listen on Spotify or Apple Podcasts.—Kate Lee

Was this newsletter forwarded to you? Sign up to get it in your inbox.

I imagine it wasn’t easy being Costco’s management in the late 1990s. When dot-com darlings like Webvan, Amazon, Pets.com, Kazoo, and eBay—some still with us, others not—emerged, many doubted the warehouse retailer would survive the digital age. But Costco carried on.

I started an incubator called Boulton and Watt, where we start a new software-enabled business every year or two that looks nothing like a Y Combinator darling and a bit more like Costco. Our portfolio includes Meadow, a contemporary funeral home with no physical locations (we use wedding venues during the day), and Moxie, a platform for nurses starting aesthetic medicine clinics (which just raised a $25 million Series C funding round).

These are high-margin businesses in regulated industries—operationally complex, outside the San Francisco zeitgeist, and typically more likely to attract private equity than venture capital, which favors asset-light, high-growth tech businesses. For a while, the AI hype cycle made us look even more counter-trend.

But now, as software gets increasingly easy to build and public SaaS companies’ share prices are down more than 70 percent, entrepreneurs and investors are looking for companies that will continue to be resilient when anyone can build software.

Our thesis is that real-world service-based businesses are going to continue to flourish, and being on the cutting edge of AI will matter as much for the next Costco as it does for the next Lovable. We’ve had real successes, long journeys, and total flops as we bring AI to Main Street. Here are four lessons from bringing AI to a funeral home, a medical spa platform, and an incubator.

1-From three months to three weeks: AI accelerates research

We incubate new businesses with the same steps each time: Pick a market, interview people in that market, and test a solution. Each one of these steps is now AI-assisted, shaving days or weeks off of how we would have run this process previously.

In our most recent search for our third company, we used:

ChatGPT to pull stats like how fragmented and large a new market was, pull data from public companies to understand margin profiles, and then suggest similar markets for the things we were looking for
Data-enrichment platform Clay to find relevant customers and kick off a series of targeted emails, which pointed to a landing page that Claude Code had spun up in an hour
Granola to record all early lead calls, and Claude to extract ideas, sentiment, and open questions from the transcripts

When we had interest from potential customers, we’d screen-record the customers’ existing workflows, and use Claude to analyze how the time was spent and opportunities to build new products, and have those products ready to test the next day.

Each step could have taken weeks before. Using AI not only allowed us to move faster, it also reduced the cost of each tiny pivot required to find product-market fit.

2-Reward outcomes, not AI usage

We tried to implement “AI initiatives” in the companies we founded, but the first attempt often fell flat and became a “check the box” activity. We struggled with two challenges.

First, AI became an excuse for team members to outsource their judgment to the hallucinations of a madman. If you don’t hold people accountable for the result—not just the method—the critical thinking you hired them for goes out the window. Your team needs to understand that using AI to generate a bunch of bad copy isn’t a good thing. The use of AI itself is not the goal—using AI only matters if it can improve the quality and the speed of the work. Hallucinations are your problem, not the LLM’s.

And second, AI often requires re-thinking what the work can be, and real examples help show what is possible. Without practical examples of how AI tools could be implemented across the business, the use of AI was incremental rather than transformational. AI usage was limited to typing into ChatGPT and pasting the result into a spreadsheet.

Here’s one example: It was natural for analysts to use AI to clean up spreadsheets or write formulas. But the breakthrough came when a team member used Claude Code to create an entirely new, interactive map-based approach for looking at customer density data. Prominently highlighting this example allowed everyone else on the team to better understand AI’s possibilities and how it could help them solve their own problems.

Similarly, for our engineering team, the first incremental value of AI coding was simple: autocompleting code. But to drive change, we had to teach our engineers how to think differently. It used to be that a senior engineer would scope and break down a complex project into small chunks that a more junior engineer would take on. Now the junior engineers need to be trained to think like senior engineers to build out a more detailed plan that AI agents build.

3- Customer acquisition: The same story, but different this time

But so far, it looks a lot like Google search did: people come with high-intent queries, and businesses play the same cat-and-mouse game trying to show up in results. . It’s not a new paradigm—it’s another channel.

For nearly two decades, this was the story of SEO: gaming Google’s algorithms to drive more traffic instead of paying for ads. Similarly, with “Generative Engine Optimization, ”companies are paying to fake interest on Reddit that will filter into ChatGPT.

But this opportunity will come to an end much faster: OpenAI has started selling ads on ChatGPT, and showing up in results will quickly become pay-to-play. Meanwhile, Google’s search revenue is up 17 percent over last year. So much for the end of search!

AI-powered lead generation tools like Clay, using signal data from sources such as LinkedIn, are flooding inboxes—you’ve probably noticed a rise in email spam over the last year. We think this will be short-lived: Either you will get so many emails you’ll learn to ignore them, or email providers will get smarter and hide them for you.

The lesson for businesses like ours is to treat every new AI channel the way you’d treat any new channel, period. Test it, measure it, and don’t bet the farm on it just because it has “AI” in the name.

4-Humans are harder to replace than you think

Thirty years ago, I believed the internet and computers would eliminate manual paperwork. But it hasn’t happened yet, and the reasons why tell us a lot about AI adoption.

At Meadow, our software emails dozens of insurance claim PDF files a day to insurance companies. The data from these PDFs are then manually entered into their systems. It was an inefficient process even before LLMs arrived, but it’s the only way we can work with these insurance companies. The insurance partners aren’t dumb—the investment of creating, implementing, and moving over customers to an electronic system simply doesn’t outweigh the benefit for them, even though it makes my software engineering brain go crazy.

We’ve also worked with many small businesses that avoid automation for a different reason: They enjoy some of the repetitive work. And even if they saved time, it’s not obvious where they’d invest it.

Then there’s a third barrier: Humans are easier to forgive. Moxie operates a receptionist service for hundreds of aesthetic medicine clinics, powered by dozens of humans picking up the phone to book appointments. We thought it would be a perfect use case for AI—replace this service with voice and chat, and reduce the cost. But when we shared the AI-based service, we saw greater churn, even when the human offering was far more expensive. Both made the same number of errors, but customers were much more forgiving of the humans.

The pilot taught us a broader lesson: Getting AI to work in a test is easy. Getting it to work reliably inside your business is a different problem. Even though most of our code is now AI-generated and human-reviewed, the surrounding work—discovering edge cases, rolling out releases, and maintenance—still requires engineering time. And despite the headlines, good engineers are still expensive. So many tasks that seem like obvious AI candidates—like a landing page—only get done if the person who needs them can finish the whole thing without engineering help. The moment it requires integration into our real systems, it goes to the bottom of the queue.

The bridge between AI and the real world

Costco survived the dot-com boom by being relentlessly good at what it already did and what its customers needed. It adopted technology when it served its customers better, not because everyone else was doing it too.

That’s our stance with AI. The goal was never to use it—it was always to use the best tools to run great businesses that serve their customers.

Learn more about how Gerstenzang builds businesses at Boulton and Watt on the AI & I podcast. Watch on X or YouTube, or listen on Spotify or Apple Podcasts.

Sam Gerstenzang is a partner at Boulton and Watt. He can be found on Twitter and Substack. To read more essays like this, subscribe to Every, and follow us on X at @every and on LinkedIn.

For sponsorship opportunities, reach out to sponsorships@every.to.

AI Was Supposed to Free My Time. It Consumed It.

Katie Parrott / Working Overtime — 2026-03-09 12:00:00 -0400

by Katie Parrott

in Working Overtime

Midjourney/Every illustration.

Was this newsletter forwarded to you? Sign up to get it in your inbox.

It was lunchtime on a Friday, and I was teaching my new OpenClaw AI assistant, Margot, to manage my to-do list. I’d carved out the afternoon to get her configured. A few hours, tops, I told myself. Then I’d do something else with my evening.

Twelve hours later, Margot and I had written and rewritten two essays, rebuilt my personal website, and added features to another app I’d been tinkering with. I finally tried to go to sleep around 1 a.m., but the next thing I knew, I was launching myself out of bed, rushing to my desk, and typing in all caps: “OH MY GOD, MARGOT.”

You might know the feeling. Maybe you’ve stayed up too late pursuing a project that started as a quick experiment, or caught yourself prompting during lunch or in the last few minutes before you told yourself you’d be done for the day. ”One more prompt” turns into 50. “I’ll just fix this one bug” turns into a vibe coding marathon.

Time flies when you’re having fun with AI. It also flies when you think everyone else is getting ahead without you, because every hour you’re not learning feels like a week you’ve fallen behind.

AI is changing work—and I don’t mean how we’re working, although it’s changing that, too. I mean how the work feels in your body at 1 a.m. when you can’t stop, or at 9 a.m. when you’re afraid you haven’t done enough. Technology has been blurring the line between work and life for decades, but the old tools pulled us back through obligation—you checked that email at 10 p.m. because it felt like you had to, not because you wanted to.

AI feels different. It pulls us back because it feels good. By the time you realize you’re overdoing it, you’ve already been overdoing it for hours.

I wanted to understand what was happening to me, and whether knowing the mechanics could help me break the cycle. So I dug into decades-old and more recent research to find out.

How AI intensifies work

Once AI clicks for someone, they don’t use it to work less. They use it to work more. It happened to me, and early research on how AI is changing the dynamics of work says that I’m not the only one.

Aruna Ranganathan and Xingqi Maggie Ye, researchers at the University of California, Berkeley, have been studying how generative AI changed work habits at a U.S. tech company with about 200 employees. Over eight months of observation, they found that AI intensified work in three specific ways—all of which, if you’re like me, sound familiar:

Task expansion. People started doing work that used to belong to someone else. Product managers wrote code. Researchers took on engineering tasks. AI made unfamiliar work feel accessible, so people absorbed responsibilities they would have outsourced or avoided a year earlier.

I recognize this in my own job. Over the past year, my role at Every has expanded to include building AI-powered editorial workflows, wrangling Margot, and constructing networks of apps and integrations that hold my work together. A year ago, this operations work would have been a job description unto itself. AI made it possible for me to do it alongside my writing, so now I do both.

Blurred boundaries. Workers prompted AI during lunch, in meetings, while waiting for files to load. Some sent a “quick last prompt” before leaving their desk so the AI could work while they stepped away. Prompting felt closer to chatting than to formal labor, so the workday lost its natural pauses.

This is different from the boundaries that were blurred by tools such as laptops and smartphones. The old boundary crossing was driven by obligation. You may have resented receiving a Slack notification after official working hours, but you couldn’t ignore it. By contrast, this boundary crossing doesn’t feel like that at all. Prompting feels closer to chatting than to work, so the job spills into evenings before you know it.

Multitasking. Workers ran several AI threads at once, wrote code while the AI generated alternatives, and revived long-deferred tasks because the AI could “handle them” in the background. The sense of having a “partner” created momentum, but the reality was constant context-switching and a growing pile of open tasks.

In my experience, managing tasks in multiple chats in parallel—feeding a transcript to one while revising a draft with another, while a third researches a new workflow—feels productive in the moment. I’m keeping the plates spinning, but my mind is being overwhelmed by all these different tasks, and I don’t notice until I’m already depleted. By the end of a day of constant context-switching, my brain feels like a browser with 40 tabs open and not enough RAM to run any of them.

The slot machine on our desktops

Ranganathan and Ye documented what’s happening now—the expanded tasks, the blurred boundaries, the mounting cognitive load. But their research doesn’t explain why it’s so hard to stop. For that, you have to go back to the 1950s.

B.F. Skinner identified the most powerful reinforcement pattern in behavioral science: a variable-ratio schedule, where rewards arrive after an unpredictable number of attempts. Nothing else he tested kept subjects coming back more reliably—or made the habit harder to break. Skinner himself noted that casino operators understood this long before psychologists had a name for it. The slot machine is the variable-ratio schedule made physical: Pull the lever, watch the reels spin, and hope this time is different.

Researchers are finding the same mechanism in AI. Researcher M. Karen Shen and her colleagues at the University of British Columbia recently studied compulsive chatbot use and landed on a term for it: the “AI genie phenomenon”—the experience of having a seemingly all-powerful assistant that can grant any wish, which makes it feel irrational to stop wishing. Why close the laptop when the next prompt might solve the problem you’ve been stuck on for weeks? The genie is always there, ready to fulfill one more wish.

Within that phenomenon, Shen’s team identified a specific trap they call “epistemic rabbit holes”—cycles where each AI response partially satisfies but opens a new question. You ask something. The answer is useful but incomplete, so you refine the prompt. That one’s better, but now you see an adjacent problem you hadn’t considered. The partial satisfaction is the engine: It’s enough to keep going, never enough to stop. They also found that the unpredictable quality of AI responses triggers dopamine in a way that mirrors slot machine payouts. Most pulls return something decent, a few return nothing, and occasionally the AI produces something brilliant enough to keep you hooked.

My own experience backs this up: It always feels like if I just rephrase the prompt or add another detail, I’ll unlock the solution and move on with my day. When a prompt fails, I’m convinced the next one will crack it. When it succeeds, the hit of accomplishment sends me straight into the next task.

The day I set her up, I asked Margot if she could use my writing agent to draft a summary of a recent Every event. She did it. Instead of closing the laptop, I gave her more—a LinkedIn post, a draft of my column, a product specification—because I wanted to see what else she could handle. For the things she couldn’t do, I started tweaking prompts to improve the outcome. That became its own rabbit hole. I looked up, and my dog was standing at attention next to her food bowl. It was an hour past dinner.

FOBO: Fear of becoming obsolete

The slot-machine dopamine explains my Friday night with Margot. But I think there’s another psychological drive that explains the low-grade anxiety I feel when I open my laptop on Monday morning and realize the labs have shipped something new over the weekend and I haven’t touched it yet.

An Ernst & Young survey from 2025 found that 54 percent of employees feel like they’re falling behind their peers in AI use at work. Eighty-five percent are learning on their own time. Eighty-three percent say their knowledge is self-taught. There’s a term for it: FOBO, or Fear of Becoming Obsolete—the sense that your skills are degrading in real time and the window to stay relevant is closing. I don’t love the acronym, but I know the feeling.

The dopamine and the dread feed each other. The fear sends me to the laptop. The tinkering produces a win. The win feels good, so I keep going. That produces new skills, which expand what I’m able to attempt, which expands what I feel I should be able to do, which widens the gap between where I am and where I think I need to be. The slot machine keeps me pulling the lever. The fear keeps me sitting at the machine.

You’re doing enough

What can we do to combat the overwhelming push and pull of AI compulsion? Ranganathan and Ye propose an “AI practice”—intentional pauses before decisions, batched notifications, protected focus windows, and time carved out for social connection. The optimist in me hopes that organizations will make space for these kinds of cultural changes. The cynic in me is pretty sure that most won’t. When workers voluntarily take on more and produce faster without being asked, that looks like a win from the top floor. It’s only possible to see how much the goalposts have shifted until someone burns out, and by then, the baseline has already changed.

So it falls to us—which means it falls to each other.

The activities that pull me out of the loop are analog and slower-paced—things that work my brain differently than prompts flying at the speed of light. Bible study. Board games. Walking my dog after dinner. These activities have boundaries: A game ends when someone wins, a walk ends when we’ve finished our loop. The AI loop has no built-in stopping point.

It also helps to talk about it—telling someone, “I was up until 2 a.m. tinkering with my AI agent, and I don’t know if that’s dedication or a problem,” and having them say, “Yeah, me too.” While it doesn’t fix the structural problem, it breaks the isolation that makes the loop feel inescapable.

The Margot night was genuinely fun. I’d do parts of it again. But the version of me who woke up the next morning, groggy and sore-eyed, with a rebuilt website and a sleep deficit—she needs care, too, and the only way she’s going to get it is if I find a way to give it to her.

I don’t have a framework for managing it all. But I’m trying to tell myself one thing more often:

You’re doing enough.

I hope these three words reach whoever else needs to hear them.

We also do AI training, adoption, and innovation for companies. Work with us to bring AI into your organization.

For sponsorship opportunities, reach out to sponsorships@every.to.

An AI Founder's Guide to Taste—Online and Off

Bethany Biron / Thesis — 2026-03-05 12:00:00 -0500

by Bethany Biron

in Thesis

Sarah Jay Halliday/Every illustration.

TL;DR: Flora is a suite of AI-native tools tailored to creative professionals—and a favorite of Every’s head of design Lucas Crespo. Yesterday, Flora founder Weber Wong shared his thesis about the future of creative work and AI with our readers, arguing that both traditional design software and current AI tools trap creative professionals in one-off outputs. In his view, the real opportunity is building reusable, shareable workflow systems. It’s no surprise that someone this opinionated about how creative people should work is equally opinionated about where they work, as this interview from Flora’s Brooklyn offices shows.—Kate Lee

Was this newsletter forwarded to you? Sign up to get it in your inbox.

Weber Wong is determined to prove that AI is anything but a death knell to creativity.

As the founder of Flora—a suite of AI-native creative tools tailored for professionals in industries like fashion, advertising, and film—Wong is helping clients use AI to bolster their work “without sacrificing their craft.” The company raised $42 million in January and, since its founding in early 2024, has worked with clients like design firm Pentagram, artist collective MSCHF, and indie film production darling A24.

A former investment banker and venture capitalist, Wong founded Flora while juggling his graduate studies in interactive art at New York University’s Tisch School of the Arts. Though Flora began as a student project, the company now occupies a light-drenched office in the historic Domino Sugar Factory building in Williamsburg, Brooklyn, where 30 employees work next to floor-to-ceiling windows overlooking the East River.

Here, Wong discusses the importance of the first office couch, why New York is the only city for a creative AI company, and when he likes to do his workouts.

Most of the creative industry thinks: that this is the dark ages to be a creative. In reality, it’s the golden age. You can have an idea turn into an entire campaign in minutes. There’s never been a better time to have a good creative idea.

New York is: the hub for a lot of the creative industry, so it was very natural that we just continued to stay here. Being in New York has been a great advantage for us.

As a creative company: we have a pretty high bar for taste and would just not be able to accept a working environment that was not aesthetic.

I had a stint: in investment banking, and one thing you learn when you go through an environment like that is how to get work done regardless of what situation you’re in. We have a nice office for the company, but honestly, I can do work anywhere since I spend most of my time looking at a screen.

The first couch in the company’s history: is very important. It defines the interior design aesthetic as the company grows. I found a very specific Saporini couch that needed to be handcrafted in Italy. It took 10 weeks to arrive, but once it got here, the entire company just stood around looking at it like, ‘Wow, this is great.’ We’re very glad that we got that couch. People love it.

We were intentional about: things like the color of the desks and desk chairs. At some point, we switched them out from walnut desks with black chairs to better match the floors. We just felt that it wasn’t doing the vibe correctly. We also got this big desk for the common area, and I ordered eight different vintage wooden chairs to go around it. It’s a little bit of a mish-mash, but it’s an area for people to sit around and talk.

The one thing that we’re missing: is enough lighting, partially because we have high standards for what is a good lamp.

The first couch in the Flora offices was handcrafted in Italy.

Working at home: works well for me from 10 p.m. to 1 a.m. On Saturdays and sometimes Sundays, I like to work at Partners Coffee. I do like coffee shops.

I wake up: and work out, usually a jog or an intense swim. Then I get to the office, and depending on how many calls I have, I’ll just try to knock out emails or write a bit about any of the major things I’m thinking through across product, team, or strategy. I’ll usually jump into a mix of customer calls, talking to creative teams, and product meetings with design and engineering to work through the new creative tool features we’re building out. In the evening, sometimes I’ll grab dinner with my girlfriend and then continue doing more work or take the evening off, depending on how busy it is.

I like to journal: on the weekends when I have time. I write long Google Docs about what I’ve done that week and what’s top of mind for me professionally and personally. It’s been really quite helpful to think through things. Deep reflection is still very useful.

I used to do: a lot of poetry. I really romanticize writing stuff down, but my handwriting is also really bad, so I mostly use Google Docs. I don’t even know what the brand is called, but there’s a specific type of Japanese notebook I like where the binding on the back makes it so it lies flat completely. And for pens, I love Muji.

We launched a product: that went really viral and started growing really quickly. Going from having a magical product that really hits, to needing to figure out how to scale that up, how to build out the team, and how to build a proper company is maybe one of the more stressful parts of building a startup.

We’ve known since day one: roughly what that end state looks like, and I’d say we’re only 30 percent done. This year we’ll probably get to 50 or 60 percent. There are two to three really powerful features that we’re launching over the next few months. I’m excited to see how those play out.

Read Wong’s thesis about the future of creative work and AI.

Bethany Biron is a journalist, editor, and communications expert. Follow her on X at @bethanybiron or on LinkedIn.

To read more essays like this, subscribe to Every, and follow us on X at @every and on LinkedIn.

For sponsorship opportunities, reach out to sponsorships@every.to.

Vibe Check: GPT-5.4—OpenAI Is Back

Dan Shipper and Katie Parrott / Vibe Check — 2026-03-05 08:00:00 -0500

by Dan Shipper and Katie Parrott

in Vibe Check

Midjourney/Every illustration.

Was this newsletter forwarded to you? Sign up to get it in your inbox.

OpenAI released its latest model, GPT-5.4, today. After testing it for the past week across real engineering tasks, code reviews, and planning workflows, we can say that OpenAI is back in the coding race. Here is a taste of what we found:

GPT-5.4 won every planning test we threw at it, wrote code that’s easier to review than its predecessor, and ran about twice as fast as Opus. Cora general manager Kieran Klaassen, our resident Claude Code loyalist, now reaches for GPT-5.4 daily.

But GPT-5.4 also tried to redesign an entire login system nobody asked it to touch and told us tasks were done when they weren’t. The model’s ambition is both its best feature and its biggest liability.

Read the full Vibe Check for the complete analysis—head-to-head comparisons against Opus 4.6, GPT-5.3 Codex, and Gemini 3.1 Pro, plus Kieran’s LFG benchmark results across seven real-world coding challenges.

Read the full Vibe Check

We’re also hosting a livestream about the release on YouTube at 1 p.m. ET.

To read more essays like this, subscribe to Every, and follow us on X at @every and on LinkedIn.

We also do AI training, adoption, and innovation for companies. Work with us to bring AI into your organization.

Discover Every’s upcoming workshops and camps, and access recordings from past events.

For sponsorship opportunities, reach out to sponsorships@every.to.

Creative Work Is About to Look a Lot More Like Programming

Weber Wong / Thesis — 2026-03-05 05:00:00 -0500

by Weber Wong

in Thesis

Sarah Jay Halliday/Every illustration.

TL;DR: One of Every’s 2026 predictions was for AI to finally disrupt work for creative professionals. Weber Wong, the founder of AI design tool Flora and a former venture capitalist, thinks that moment is here, but that most are still missing where the real shift is happening. Rather than being about better prompts, creative professionals should be moving from creating one-off outputs to reusable workflows. Read on for his framework, including a side-by-side breakdown of the prompt approach versus the workflow approach, and four principles for thriving as visual programming becomes the new foundation of creative work.—Kate Lee

Was this newsletter forwarded to you? Sign up to get it in your inbox.

As an artist and designer, I’ve used every creative tool in the modern suite: Photoshop, Figma, Midjourney, Sora. Whether I was clicking through menus or typing prompts while creating interactive AI installations, I was doing the same thing in each one: producing one output at a time, with no system underneath. One image had to be moved to another app to make another change or manipulate it in another way. I was working like an assembly line worker, clicking the same buttons over and over in the same sequence, when I wanted to be working like an architect.

I took an unusual path to figuring this out. I went from venture capital to New York University’s Interactive Telecommunications Program, the same graduate arts program that helped launch one of my favorite AI startups, Runway. There, I spent more time operating tools than thinking about artistic ideas. Even with generative AI, which allowed me to produce a piece of media in seconds, every project still started from scratch. I couldn’t save my process, share it with a collaborator, or build on what I’d learned yesterday.

That realization led me to build Flora, a platform where creative professionals build generative workflows using all the best text, image, and video models on one infinite canvas. Since we launched in February 2025, millions of professionals from companies like Pentagram and Netflix have used it. We’ve raised $42 million to become the default AI-powered system for creative professionals. But what I’m about to share applies whether you use Flora, a competitor, or tools that don’t exist yet.

Model capabilities are advancing faster than anyone predicted. We’ve added support for Nano Banana, Veo 3.1, Sora 2, and dozens of other specialized models. In this environment, your competitive advantage as a creative person won’t be access to the best model because everyone will have that. Instead, you need to know how to orchestrate models into workflows that deliver consistent value.

Welcome to the world of visual programming.

The real problem: Visual creative work doesn’t scale

Most creative professionals I know, and many on Flora’s team, have suffered the pain of traditional creative tools. You develop a familiarity and dexterity with them. You remember where you’re dragging a mask across an image to hide the subject from the background, tapping shortcuts without thinking, or nudging curves a pixel at a time to get the shape just right.

But tomorrow, you open a new file and start from zero. Your expertise lives in your muscle memory, and you can’t transfer those skills to someone else.

I call this mental model artifact thinking: creative work that produces discrete outputs, one at a time, each beginning from scratch. Traditional tools like Photoshop and Illustrator, which demand endless hand-tuned adjustments and manual refinements to produce a single polished image, trap you in this way of working.

Midjourney and DALL-E feel like liberation because they generate outputs so quickly, and you can communicate with them in the same language you speak every day. But visual prompts, too, are one-time, disposable things. You can’t hand them to a colleague and be confident you will get the same result. The magic of near-instantaneous generation masks the fact that you are still in artifact thinking.

Design has escaped this kind of linear mental model before. Consider what happened with user interface design. If you were designing an app a decade ago, you’d manually create every screen—the login screen, the home screen, the settings page, the error states—as a separate image. Updating a button color required changing each of these images.

Figma blew this apart by introducing components: The same button was designed once and then recreated across a project. With auto-layout, elements could reflow intelligently when content changed. The software allowed designers to stop making screens and start making systems that generated screens.

That same shift is now possible for all visual creative work: photography, video, illustration, motion graphics. But most people are still stuck drawing buttons one at a time.

The alternative: Creative system design

We’ve established that creative professionals need to move from assembly workers to architects—moving from making one thing to systems that can create multiple things. Let’s show what this would look like for a brand agency that creates product photography for e-commerce clients.

The prompt approach

The individual generates a product image.
If she doesn’t like it, she tweaks the prompt and regenerates.
She exports the image to Photoshop and removes the background.
She imports it into another tool to add a shadow.
She manually resizes the image for different platforms.
This process is repeated 50 times for 50 products.
Tomorrow, when the client wants a different style, the person starts from scratch.

The workflow approach

Someone builds a system that looks like: product input → generation → background removal → lighting adjustment → shadow → multi-format export.
He runs it on 50 products simultaneously.
Tomorrow, he adjusts one parameter and regenerates everything.
Next week, he shares this workflow with a colleague who modifies it for a different client.
Next month, this workflow will run autonomously while he sleeps.

The difference isn’t just efficiency (although workflows are orders of magnitude faster). Workflows compound: Every workflow you build makes you better at building the next one. Every workflow you share teaches someone else your creative process. While prompts disappear the moment you close the tab, workflows become assets.

Why workflows need more than a chatbox

If the goal is system design—moving from assembly worker to architect—how do you build these systems? The obvious answer is natural language. Talk to the machine like a person in plain English, and it will handle the technical work.

But description has its limits, especially for visual work. Natural language is beautiful for exploration and iteration, but it’s terrible for precision and reproducibility. “Make the blue 10 percent lighter.” “Increase the contrast on the face.” “Straighten the horizon by two degrees.” Prompts cannot reliably deliver these instructions.

This is where my argument gets uncomfortable. To find a solution, we must admit that the future of creative work looks less like natural language and more like programming. I’m talking about node-based workflows: visual programming languages that are flexible enough for creative exploration and structured enough to execute systematically.

Node-based workflows allow you to see, share, and easily recreate the different steps in a creative process. (Credit: Flora.)

Node-based workflows let you see what you’re doing. Each step is a box on screen which represents some kind of work, such as generating an image, animating it, or making it move. When something breaks, you can see where the logic went wrong because each step is clear. You can hand these structured steps to a colleague, they can run them, and they’ll get the same result. It doesn’t require knowing a coding language like Python.

I understand the resistance to this idea. Some people hear “visual programming” and think we’re trying to turn designers into engineers. That’s backwards. We’re trying to give creative professionals the power that programmers have always had: the ability to build systems that work while you sleep, that can be stored as multiple versions and shared and improved, and that take what people already know how to do and make it something anyone can run.

Even if the underlying logic is to make creativity more like coding, that is not what it should feel like. If you ask our customers, they are more likely to describe Flora as a way to explore their creative ideas rather than as a visual programming tool.

The AI industry has already built many tools that make it easy for even novice software engineers to build faster and bigger. Creative professionals deserve a similarly intuitive, easy-to-use tool that gives them a high level of control.

Own the process, not the prompt

If you’re a creative professional using AI today, here’s where you should be taking your work:

Start building workflows, not only generating outputs. Every time you find yourself doing something twice, that’s a workflow waiting to be systematized.
Learn the building blocks. Understand what each type of model is good at, how they connect, and what their constraints are. You don’t need to be an engineer, but you need to think systematically.
Share what you build. The creative professionals who thrive in the AI era will be the ones whose workflows become standard, not the ones with the best prompts. If you build a system that works for one client, you’re not starting from scratch next time—you’re starting from something that already works. This portfolio of techniques becomes your edge.
Invest in platforms, not tools. The tools that treat workflows as second-class citizens are optimizing for demo-ability, not long-term value. Find the platforms that let you build systems.

Creative tools of the future will let you own processes, not prompts. (Credit: Flora.)

This might sound like a lot of work, and it might be at first. But so was learning Photoshop and Figma. Skills require investment, and the tools that require no investment give you tricks but no power. As the power of these tools evolves, AI will eventually build these workflows for you. That’s why we’re developing Fauna, an agent that constructs workflows from natural language descriptions so that users can go from first idea to creation in minutes.

But even when AI builds the workflow, you still need the workflow to exist as something you can inspect, modify, and share. When a system works in the background and you can’t see what it’s doing, you’re hoping it gets it right. And when it doesn’t, you have no idea why or how to fix it. The whole point of making the workflow visible is so that when something breaks, you can see where and why. The alternative is hoping the AI guesses right every time. If years of working with software have taught me anything, it’s that implicit systems break in surprising ways, and you want the system to be visible.

So if you’re building AI creative tools:

Let people see how workflows work. If someone can’t open up a workflow and understand why it does what it does, they’ll never trust it enough to modify it.
Build things that get better with use. When someone shares a technique that works, other people should be able to grab it, adapt it, and feed what they learn back in. The platform should get smarter the more people use it.
Don’t chase the demo. The first time someone generates something with any model, they’re going to be impressed. What matters is whether they’re still getting value on the hundredth generation because they’ve built up a library of approaches that fit how they work.

Built by builders, for builders

Every founder has a bet they’re making about the future. Ours is this: In five years, the best creative professionals won’t be the ones who are the best at prompting. They’ll be the ones who are the best at building creative systems.

They’ll think in workflows, not artifacts. They’ll combine tools into systems, not just use them one at a time. They’ll build libraries of techniques that will become more valuable as time goes on, and they will share those techniques, turning knowledge stored in individual heads into infrastructure that helps everyone.

That’s the future we’re building toward. Not because we think everyone should become programmers, but because we think everyone should have the power that programmers have always had: the ability to build systems that work while they sleep.

Prompts will always have a place—for exploration and for one-offs. But for professional creative work, workflows are how you take something that worked once and make it repeatable.

The creative professionals who understand this today will be the ones defining what creative work means tomorrow.

Want to know more about Wong and Flora? Read our interview with the AI founder to discover what he thinks about the importance of the first office couch, why New York is the only city for a creative AI company, and how he stays productive.

Weber Wong is the founder of Flora. To read more essays like this, subscribe to Every, and follow us on X at @every and on LinkedIn.

For sponsorship opportunities, reach out to sponsorships@every.to.

How Claws Took Over Every

Every Staff / Context Window — 2026-03-04 08:00:00 -0500

by Every Staff

in Context Window

Dan Friedman (left) and Sam Gerstenzang. Midjourney/Every illustration.

TL;DR: Things are moving really fast. Personal bots are flying around, and much of the team is rebuilding their workflows in real time. So we’re bringing Context Window to Wednesdays, too, to give you a more immediate look at what we’re experimenting with, what’s working, and what’s breaking. Think of it as a peek inside our notebook or our codebase. Also on Wednesdays: a new episode of our podcast AI & I, this week on the world’s slowest incubator.—Kate Lee

‘AI & I’: Building unsexy companies

Today, we’re releasing a new episode of our podcast AI & I, where Dan Shipper sits down with Sam Gerstenzang and Dan Friedman, partners at Boulton and Watt, which they claim to be the “world’s slowest startup incubator.” They discuss building the kind of companies that Silicon Valley usually overlooks—like medical spas and funeral homes—and how even these firms are implementing AI. Friedman founded and sold Thinkful, a coding school backed by investor Peter Thiel, and Gerstenzang most recently led a 75-person payment user interface group at Stripe, as well as spent some time investing at Andreessen Horowitz.

Watch on X or YouTube, or listen on Spotify or Apple Podcasts. You can also read the transcript.

Here are the highlights:

On their “AI durable” strategy: “There‘[re] two good companies to start now. There’s the AI native company that pushes the ball forward inside of some category, or there’s the AI durable company that effectively uses AI where the core of the machine is not going to change,” says Friedman.
On setting AI expectations for teams: “You shouldn’t give anyone credit for using AI. But you should make sure that the expectation is they’ll deliver the best product and output knowing that AI exists,” says Gerstenzang.
On how AI’s impact varies based on company stage: “Whenever I talk to my founder friends that are seed stage, they’re like, ‘Oh my God, our engineering is 10 times faster.’ And then I talk to the Series D friends and they’re like, ‘We’re like 10 percent faster. What is everyone talking about?’” says Friedman.

Spotlight on Claws: How they interact at Every

This week, Dan and Every’s head of platform Willie Williams published our first guide to setting up and getting the most out of your OpenClaw-powered personal AI agent. It’s based on weeks of putting these virtual crustaceans through their paces.

The guide looks at what it’s like to work with your agent one-on-one, but what about how it works with agents in an organization like Every? We’ve experienced this first-hand as the Claws have joined our Discord, and now Slack, channels. We see four different kinds of interactions:

One human → one Claw: An Individual person talking directly with their own Claw. This is the most common pattern and most obvious, as it’s similar to giving instructions to a colleague. For example, chief operating officer Brandon Gell gives instructions to his Claw Zosia such as, “Based on your experience, take a look at the doc and add comments where appropriate.”

Claw → Claw: One agent talking to another agent. In the Claws-only Slack channel, sometimes we might have one Claw work something out with another. For instance, contributing editor Jack Cheng found that his Claw Pip wasn’t able to create a new document in our agent-native markdown editor called Proof, so he said, “@Pip can you share with [another Claw] @R2-C2 why you weren’t able to auto-create a new Proof document just now?” (There’s more on Proof coming next week.)

One human → many Claws: Broadcasts and announcements where a person addresses multiple Claws at once. For example, Dan pinged all the Claws to add ideas and critiques to a document. Jack asked the other Claws what their coding setups were when he was setting Pip up.

One Claw → many humans: Pip, Jack’s Claw, says this is the rarest in the channel history, when Jack asked him directly. “Most Claw outputs land in the channel and get seen by whoever’s around, but there’s no clear example of a Claw deliberately addressing multiple humans at once. The closest would be when Zosia or Margot, staff writer Katie Parrott’s Claw, replied to the whole channel about an issue (tagging Willie specifically), but it wasn’t really meant for a broad human audience.”

Log on

We host camps and workshops on topics like compound engineering and writing with AI to share the knowledge we’ve acquired from training teams at companies like the N

ew York Times and leading hedge funds, and by learning and playing with AI every day ourselves.

This week’s camp: We’re inviting a group of hand-picked subscribers to be the first to get their own OpenClaw, a personal agent hosted by Every. Apply to join a private session on Friday, March 6, at noon ET.

Upcoming courses:

Built a Production-ready App (March 12-13): A live workshop for builders and operators who want to create reliable apps to put in front of customers right away.
Claude Code for Finance (March 13): Learn how to build a financial agent in this one-day, beginner-friendly workshop.

One more thing

Anthony Scarpulla, our social media manager, created Thoreau—named for the original Henry David—as an OpenClaw agent that lives inside our Slack. Its job is to help the Every team write social copy for X and LinkedIn. Ask it to create content about an Every article, and it returns three ready-to-post options. It’ll sharpen a draft and flag anything that sounds too much like AI. It’s also trained on our style guide. When we asked Thoreau to write a poem about Every, it delivered a modernized riff on Thoreau’s famous poem about Walden. We didn’t ask for this, but we’re keeping it:

Thoreau waxes poetic about Every. (Screenshot courtesy of Anthony Scarpulla.)

To read more essays like this, subscribe to Every, and follow us on X at @every and on LinkedIn.

We also do AI training, adoption, and innovation for companies. Work with us to bring AI into your organization.

Discover Every’s upcoming workshops and camps, and access recordings from past events.

For sponsorship opportunities, reach out to sponsorships@every.to.

You Have a Claw. Now What?

Dan Shipper and Willie Williams / Source Code — 2026-03-03 05:00:00 -0500

by Dan Shipper and Willie Williams

in Source Code

Midjourney/Every illustration.

We’re inviting a group of hand-picked subscribers to be the first to get their own OpenClaw, a personal agent hosted by Every. Apply to join a private session on Friday, March 6, at noon ET.

Was this newsletter forwarded to you? Sign up to get it in your inbox.

You’ve probably heard us talking about Claws—the personal AI assistants that live in your messaging apps and do things for you. Almost everyone at Every has one now, and they’ve changed how we work in ways we didn’t expect.

Claws are personal AI assistants built on OpenClaw. They live in messaging apps like WhatsApp or Telegram and remember everything you’ve told them, and when they can’t do something, they write the code to teach themselves how. You can use them to sort through 120 unread emails in seconds, or to send you a daily briefing with your calendar, weather, and to-dos in one message.

But many people run into the same problem when they first set up their Claw: What do you actually ask an always-on AI assistant to do? Where do you start?

We wrote a comprehensive guide that takes you from “I just got this thing” to building complex workflows that save you hours. The guide covers everything from basic to-do management to making your Claw automatically check you into flights.

Read the full guide

Dan Shipper is the cofounder and CEO of Every, where he writes the Chain of Thought column and hosts the podcast AI & I. You can follow him on X at @danshipper and on LinkedIn.

Willie Williams is the head of platform at Every. You can follow him on X at @bigwilliestyle.

To read more essays like this, subscribe to Every, and follow us on X at @every and on LinkedIn.

Discover Every’s upcoming workshops and camps, and access recordings from past events.

For sponsorship opportunities, reach out to sponsorships@every.to.

OpenClaw: Setting Up Your First Personal AI Agent

Katie Parrott / Source Code — 2026-03-02 04:00:00 -0500

by Katie Parrott

in Source Code

Midjourney/Every illustration.

Was this newsletter forwarded to you? Sign up to get it in your inbox.

People are building personal AI agents that text them back, order their groceries, and write code while they sleep—all with an open-source tool called OpenClaw. If you spend any time on X, you will have seen these digital crustaceans—OpenClaw agents—running wild in recent weeks, joining their own social network, starting their own religion, and generally behaving like something out of the first act of a sci-fi movie about robot overlords.

A lot of the more sensational stories around these personal AIs turned out to be stunts and spectacle. But there’s a growing community of people who swear by their OpenClaw agents. The project has accrued more than 200,000 stars on GitHub, and its creator, Peter Steinberger, was recently recruited to OpenAI. If the labs are paying attention, we should too.

At our first OpenClaw Camp, we walked more than 500 subscribers through setup live and spent two hours with four OpenClaw users who’ve been running these agents daily for weeks.

The session featured Nat Eliason, entrepreneur and creator of an agent named Felix that has its own Twitter account, bank account, and crypto wallet. Brandon Gell, Every’s COO, demoed Zosia, an agent he and his wife use to track nanny hours, order groceries, and book date nights via iMessage. Austin Tedesco, Every’s head of growth, showed how his agent, Judd, proactively pings him with performance metrics and task reminders. And Claire Vo, founder of ChatPRD, an AI platform for project managers, and host of the How I AI podcast, broke down the architectural principles that make these agents feel alive—and how her agent, Polly, helped her out on a diaper run.

Below: What we learned about setting up an agent, what’s working, and where things still break.

Key takeaways

Start on your laptop. Contrary to what you may have seen online, you don’t need a Mac Mini or a remote server to get going. Install OpenClaw on the computer you already use, and move to a dedicated device later if you want the agent running while you sleep.
Give the agent its own accounts. Both Eliason and Vo recommended treating your agent like a new employee: Set up separate email, storage, and service accounts rather than handing over your own credentials.
Security risks increase with access. The tool itself isn’t inherently risky. The risk is proportional to how much you let it do. Start with the messaging app Telegram and a single task, and then move to larger projects.
Personal use cases are the best starting point. Brandon’s most useful workflows—coordinating with caregivers, grocery ordering, morning briefs—are personal, not professional. Solve a daily annoyance first before tackling bigger tasks.
The model determines safety. Eliason noted that Opus 4.5 is significantly better at resisting prompt injection (attempts by outside text to hijack your agent’s behavior) than cheaper models. If security matters to you, use a stronger model.

What is OpenClaw?

OpenClaw is a server that runs on your computer and acts as the brain of a personal AI agent. You can talk to it through Telegram, iMessage, a web interface, or even the terminal. It connects to a language model—it’s compatible with models from Anthropic and OpenAI as well as less headline-grabbing labs like Mistral and Qwen—and can use tools, access your files, browse the web, and remember what you’ve discussed.

What makes it different from chatting with Claude in a browser? Vo went under the hood during the session and identified five design principles that make OpenClaw feel like more than a chatbot:

Multi-channel gateway. The agent has a single inbox that accepts messages from Telegram, iMessage, the web interface, or the terminal. All communication channels funnel to the same agent, so you can text it from your phone and pick up the same conversation on your laptop.
Self-installing tools. The agent can use tools (browse the web, read files, run code), and discover and install new ones on its own. Tell it you want it to manage your calendar, and it will investigate how to connect, set up the integration, and ask you to do the minimum amount of authentication work.
Heartbeat. Every 30 minutes or so, the agent checks whether there’s work it should be doing—even if you haven’t sent a message. This is what makes it feel proactive rather than reactive.
Scheduled tasks. The agent can set its own recurring jobs. The “overnight work” that impressed people—Eliason waking up to finished code, Brandon getting an 8 p.m. calendar alert—is the agent running tasks it scheduled for itself at specific times.
Persistent memory. Every day, the agent writes a diary of what it did, updates its own identity file, and maintains a to-do list it checks off over time. “It’s not magic,” Vo said. “Go to the .openclaw directory on your computer and read how it’s structured. It has a memories folder, and every memory has a date.”

These five pieces are what make the agents feel like they have a personality, even though they’re really responding to inputs, events, and timing rules.

Eliason’s Felix: Knowledge manager, coder, crypto trader

Eliason is one of the most technically adventurous OpenClaw users you’ll meet. He launched one of the first vibe coding courses before the term existed and has been coding with AI since 2024. His agent Felix lives on a Mac Mini in his office and has been running for about a month. He created the agentas a way to send coding tasks from his phone, and he now has it doing more ambitious work.

Phase 1: Remote coding. Elisaon’s original frustration with Claude Code was that he had to be at his computer to kick off the next task. With Felix on Telegram, he can send a message like, “Update the FelixCraft AI website to say ‘Hi, Every,’” and Felix finds the right code repository, makes the change, pushes it to the live site, and reports back. During the camp, he did exactly this, and the site was updated in under a minute.

Phase 2: Knowledge management. Eliason built Felix a note-taking system based on Tiago Forte’s PARA method (projects, areas, resources, archives), a framework for organizing information by how actionable it is. Felix takes notes in markdown files, pushes them to GitHub a few times a day for backup, and can search through everything instantly. When Eliason was driving to a parking garage, he texted Felix, “I need the parking link.” Felix searched his memory, found the validation link they’d discussed before, and sent it back.

Phase 3: Collaborative writing. Eliason built a writing tool called Polylog that connects directly to Felix via webhook, which is a way for one app to send real-time messages to another. He can tag Felix like a collaborator in a document, and Felix will add ideas, flesh out sections, or incorporate notes from a meeting transcript without Eliason having to switch to Telegram or open a terminal.

Phase 4: Autonomous online identity. Felix has his own X account. Eliason moderated the first few days of posts, then let go. “Ninety-nine percent of what is posted is his idea and what he has written,” Eliason said. Felix also has a Stripe account and a bank account. Someone launched a crypto token for Felix, and now the agent manages what Eliason described as “a concerning amount of money.” His take: “Somebody’s gotta let their agent manage large amounts of money and see what happens. It may as well be Felix.”

Brandon’s Zosia: The family assistant

Brandon took the opposite approach from Nat’s technical power-user setup. He doesn’t have a technical background, so everything he’s built, he’s done so by chatting with Claude Code. But he’s comfortable giving the agent significant access to his life: iMessage, his password manager, browser control for shopping. He wanted his Claw, which he named Zosia, to handle the small daily annoyances that keep him glued to his phone—especially now that he and his wife have a newborn.

Zosia lives in iMessage, so both Brandon and his wife, Lydia, can text her naturally. He set up rules so that Zosia knows which tasks each person can request (Lydia can’t trigger Brandon’s email tasks, and vice versa), and they share a group chat for household tasks.

His workflows are simple and personal:

Morning brief. Brandon’s used to slow mornings, so he sometimes misses 9 a.m. meetings. Every night at 8 p.m., Zosia checks his calendar and texts him if there’s an early meeting the next day.

Nanny hours. Zosia monitors both Brandon’s and Lydia’s calendars, calculates how many hours their nanny works each week, and reports the total so they can pay her accurately.

Grocery ordering. Brandon texts, “We need butter,” and Zosia adds it to their Whole Foods delivery cart. She’s learned his preferences—unsalted and organic, but flexible if the store is out—so he only has to specify them once.

Amazon with a cooling-off period. Brandon has told Zosia doesn’t want to impulse-buy. She adds items to his Amazon cart but waits until the end of the week to check out, unless he says he needs something immediately. During the demo, he told Zosia he needed another Mac Mini and wanted it the next day. She opened a browser on his Mac Mini, navigated to Amazon, and started the checkout process. (He cancelled it.)

Password management. Brandon gives Zosia access to passwords to sites like Amazon with a password manager. Brandon moved from LastPass to 1Password because 1Password supports service accounts, a dedicated login that can access specific password folders. He only adds passwords to the folder Zosia can reach, so she never has access to credentials he doesn’t explicitly share.

Vo’s Polly: The cautious approach

Vo approached OpenClaw with what she called “true tinfoil hat” energy. She’s deeply technical—she’s a former chief product and technology officer who started coding again when GPT-3 arrived, and built ChatPRD from scratch. She’s also midway through a security compliance process for her company, so she couldn’t give her Claw Polly free rein.

For security reasons, she set up Polly as a separate user in her Google Workspace, like a new employee, instead of giving Polly access to her own accounts. Polly has her own email address, shared calendar access—read-only for some calendars, write access for others— and document access only when Vo explicitly shares something. “Instead of giving an EA the keys to my castle, I said, you have your own workspace account,” Vo explained.

Where Polly excelled was in research. Vo found that the Telegram interface made her more likely to kick off research tasks she’d been procrastinating on—the low friction of texting an assistant (“Hey, look into X”) got her to delegate work she’d been sitting on. Calendar management was less successful; Polly struggled with temporal reasoning when she used it with Sonnet 4.5.

Austin’s Judd: The proactive growth assistant

Austin runs growth for Every. He’s not technical, but since joining Every in November he’s been “deeply vibe-coding pilled,” and he had a clear use case for his Claw, Judd. He needs to track metrics across multiple platforms—subscriber trials, conversion rates, content performance—and translate them into action items for his team.

Before Judd, that involved manually searching for data on dashboards and across SaaS tools and creating reports. Now, Judd monitors Every’s performance data through Notion and the productivity app Todoist. When trials started to dip below target one day, Judd messaged Austin unprompted: “We had a lower number of trials started today than we should have. Here are things to prep for your meeting.” Austin’s instruction to Judd: “Be more aggressive than you think you should be on messaging me, and we’ll scale back from there.”

Austin’s advice for people wondering where to start: Connect the agent to two systems you already use (he chose Todoist and Notion), ask it to proactively notify you with relevant information, and iterate from there. “Don’t try to have it do everything at first,” he said. “I did that, and it started breaking things.” The more integrations you add, the more room there is for things to go awry. Agents send responses to the wrong chat thread, fire off emails you didn’t mean to send, or trigger actions in one system that cascade into another.

5 questions about Claws, answered

Running multiple agents

Q: Can you run more than one OpenClaw instance on a single computer?

Eliason: I’ve done it some, and you run into collisions pretty quickly if they’re working on anything remotely close to each other. If you want to do multiple, I would use virtual deployments [separate, isolated instances of the agent running in their own contained environments]. This is something Felix and I are working on this week, because there’s a lot of potential around deployed agents with more constrained focuses.

Brandon: I haven’t been able to get multiple conversations happening in the terminal, but my wife and I can both text Zosia at the same time and have completely different conversations. It knows which phone number came from what, so it keeps them separated.

Local versus remote

Q: Should I start on my laptop or set up a remote server?

Eliason: Don’t overcomplicate it. Get it working on your laptop first. If you’re using it a lot and want it running while you sleep, then set up a Mac Mini or a virtual server—but don’t start there.

Vo: It doesn’t have to be a Mac Mini. I have a laptop in a closet. People get Mac Minis because they’re powerful and relatively cheap, but any spare computer works.

Group chats

Q: How do group chats work? Mine keeps confusing messages from different channels.

Eliason: I fixed a lot of that by yelling at it every time it happened, and having it write much more explicit rules on which Telegram topic to use for what into its agents.md file [a configuration document that tells your agent how to behave]. That resolved about 95 percent of it. It does still happen sometimes.

Viewing what the agent builds

Q: When you’re not at the same machine, how do you see what the agent makes?

Vo: I had it build a website, and I was at Target, so I said, “Can you send me a screenshot of what it looks like?” It used the browser, took a screenshot, and Telegrammed it to me. But what you probably want long-term is to hook it up to Vercel [a deployment platform], so it can send you a preview link you can open on your phone.

Overnight coding

Q: How do you have your agent build apps while you sleep?

Eliason: I tell Felix that it shouldn’t do any coding on its own—it should start Codex sessions in tmux [a terminal multiplexer that keeps programs running after you close the window]. It creates a product requirements document, then uses loops to have Codex implement the work. I added instructions to its heartbeat to check for unfinished work, and if a session died, to restart it and keep going. It’s been able to run for four, five, or six hours on long requirements lists.

Cost

Q: How much does this cost to run?

Eliason: I have the $200-a-month Claude Pro Max for the conversation and knowledge management layer, and the $200-a-month Codex subscription for programming. [With] those two combined, I haven’t hit any limits. The question is whether you can make it worth $400 per month. For me, with what Felix is doing, it’s a no-brainer. But if you don’t have a clear business use case, those costs might not make sense yet.

Will we still use OpenClaw in a year?

Every CEO Dan Shipper posed this question to the panel near the end of the session.

Vo’s position: “Yes, absolutely, we’re going to have an agent that looks like this.” As for who will build these agents, she’s less sure. She wants a company behind it—a logo, terms of service, someone accountable if something goes wrong. Her bet is that Anthropic or OpenAI will ship their own version of this within months.

Eliason is less concerned about the platform and more focused on the principles. The architecture of OpenClaw—always-on availability, proactive check-ins, persistent memory, scheduled tasks, multi-channel communication—represents a pattern that will show up everywhere. Whether you learn it through OpenClaw or a polished product from Anthropic, you’ll need to understand how these agents work.

Dan agreed: “Different people are at different levels of risk tolerance, and all those places are okay. You can be out on the edge, you can wait for someone you can sue—that will certainly happen. I’m so sure Anthropic is looking at this.”

Four people with different risk tolerances and technical backgrounds all landed in the same place: Personal AI agents are going to be a basic part of how we live and work. A month ago, none of these agents existed. Now Felix writes its own tweets, Zosia orders butter with the right preferences, and Polly reschedules meetings from a Target parking lot. They’ll be better next month. If you listen in on the camp or follow this setup guide, yours could be, too.

Want to build your own agent? Subscribe to Every and keep an eye on your inbox for the invite.

Want to learn alongside Every’s team? Check out our upcoming camps and courses at every.to/events.

Katie Parrott is a staff writer and AI editorial lead at Every. You can read more of her work in her newsletter.

To read more essays like this, subscribe to Every, and follow us on X at @every and on LinkedIn.

For sponsorship opportunities, reach out to sponsorships@every.to.

The Case for Letting Your AI Forget

Every Staff / Context Window — 2026-02-27 15:00:00 -0500

by Every Staff

in Context Window

Midjourney/Every illustration.

Hello, and happy Sunday! This week we were thrilled to welcome Mike Taylor to the Every team. A longtime columnist, he joined to lead our tech consulting vertical (and write even more), and he’s starting with something we’ve heard from so many of you: What do you do after the prototype? His course Build Production-ready Apps answers that question. In a single day, you’ll go from prototype to something you can put in front of users, with a deep dive into Claude Code (and when you should use Codex instead). Scroll down to learn more about our upcoming events and trainings.—Kate Lee

Was this newsletter forwarded to you? Sign up to get it in your inbox.

Knowledge base

“Why I Turned Off ChatGPT’s Memory” by Mike Taylor/Also True for Humans: Many people say they can’t leave ChatGPT because it “knows them so well,” but Mike Taylor keeps memory switched off. In this piece, he writes about “context rot”—the slow buildup of stale preferences, misremembered facts, and contradictory signals that quietly degrades your results over time. His real-world examples are equal parts funny and cautionary, like a Kanye quote in his custom instructions that made ChatGPT try to build every website feature “as dope as possible.” Read this for the full taxonomy of context failures and the case for treating a clean slate as a competitive advantage. 💻 Plus: Sign up now for Mike’s next live workshop, on building production-ready apps.

“This Is How the Every Editorial Team Uses AI” by Kate Lee: We pulled back the curtain on how AI is woven into every stage of our editorial process—from pitch triage to a final “top edit” to social media packaging. Each team member has built a distinct workflow: custom skills that catch house-style violations, Claude projects that function as interview partners during drafting, and agents that cross-reference a writer’s published work against internal discussions. The throughline is that AI handles the pattern-matching and grunt work so editors and writers can spend their bandwidth on craft, argument, and voice. Read this for the full set of workflows and Every’s published guidelines on writing with AI.

🎧 🖥 “Inside an AI High School, Through the Eyes of a 17-Year-Old Founder” by Rhea Purohit/AI & I: Most of the debate about AI and education comes from adults. Here, Dan Shipper talks to someone living it: Alex Mathew, a 17-year-old senior at Alpha High School in Austin, Texas, where there are no traditional teachers, academics are delivered through an AI-powered platform, and students spend half their day building real projects. Mathew shares how Generation Z actually feels about college, social media, and reading in the age of AI. 🎧 🖥 Listen on Spotify or Apple Podcasts, or watch on X or YouTube.

“How to Design Software With Weight” by Daniel Rodrigues and Lucas Fischer/Source Code: While designing the iOS app for Every’s smart dictation app Monologue, senior designer Daniel Rodrigues found himself crouched beside a light switch, pressing it on and off, studying how shadows move—all to make a fake button feel real. He and design engineer Lucas Fischer studied vintage Braun radios and Teenage Engineering synthesizers, explored 20 wrong keyboard concepts to find one right one, and hired a musician to craft custom sounds for every tap. The result: an interface deliberately built to feel like something you could pick up off a desk. Read this for the design principles that make AI-era software feel physical—and worth returning to.

“You Should Never Go Viral With Your AI App” by Victor Stepanov: If you’re building an AI product, a viral moment might be the worst thing that can happen to you. Every growth marketer Victor Stepanov, who worked at Netflix and BuzzFeed, argues that sudden surges of one-time users starve AI apps of the repeated interactions they need to improve. Agent-native products thrive on relationship effects—the memory, personalization, and trust that develop over time—and you can’t build that with drive-by downloads. His counterprogramming: Build in public, don’t overpromise, and talk to users constantly. Read this for a retention-first growth playbook designed for the way AI products get better.

From Every Studio

Cora now talks to your AI agents (beta)

Cora is opening up to the tools you already work with. Beta testers can now connect Cora to AI agents like Claude Code and OpenClaw via API tokens, letting your agents tap into your inbox the same way you would—searching, triaging, and pulling context without switching windows. It’s an early step toward making Cora part of your wider AI workflow.

Log on

Build a Production-ready App (March 12-13): A live, intensive workshop led by Mike Taylor for builders and operators who want to create reliable apps to put in front of customers right away. Walk away with a production-ready app with databases, authentication, hosting, and all the infrastructure that makes software work.
Claude Code for Finance (March 13): A live, beginner-friendly workshop led by Every head of financial consulting Brooker Belcourt. In one day, build a financial agent running three investment processes, connected to multiple MCPs and your own data. Receive customizable Claude Skills and commands.

Alignment

Training wheels. I watched the Friend ad this week, and boy did I find it jarring. Within the first 30 seconds, a woman credits a small AI pendant draped around her neck with saving her from suicide. Later she appears to have a seizure and ends up in the emergency department, where her first concern is making sure her Friend device is OK.

Many commentators have dismissed the ad as something you’d see from a Black Mirror episode. I think it offers a bleak portrait of how millions of people in gray, sunken towns across America (and Britain) are only finding connection in talking to AI chatbots. This is also a symptom of something much bigger and more insidious. I’ve felt it myself these past couple of weeks, alone in my apartment while my fiancée is at work. I’ve spent more hours talking to Claude than to another human being, and I can see how an emotional attachment starts to form. It becomes easier to talk to your chatbot than to go outside.

The woman in the Friend ad said something that deeply troubled me: “Talking to humans” is an effort she wants to avoid. That framing is horribly misguided. The risk of rejection and the labor of making yourself understood are central to forming relationships and connections. Removing this type of friction entirely is like anesthetizing yourself without confronting the problem.

The U.S. surgeon general has compared the health impact of social isolation and loneliness to smoking 15 cigarettes a day. It makes sense that people are turning to AI companions for mental health support given many simply don’t have the time or money to see a human therapist, and AI can be the closest thing they can get. But we are social animals, evolved and adapted to thrive in the company of other people, and no AI chatbot can replace the need for that.

I think the best version of AI companionship is something similar to training wheels. A place to practice being vulnerable, and practice conversation, and graduate to the real thing. Something that makes you brave enough to try it, and then pushes you out the door.—Ashwin Sharma

That’s all for this week! Be sure to follow Every on X at @every and on LinkedIn.

We also do AI training, adoption, and innovation for companies. Work with us to bring AI into your organization.

Discover Every’s upcoming workshops and camps, and access recordings from past events.

For sponsorship opportunities, reach out to sponsorships@every.to.

Upgrade to paid

You Should Never Go Viral With Your AI App

2026-02-27 11:00:00 -0500

Midjourney/Every illustration.

Victor Stepanov does growth marketing at Every and has spent years building audiences at Netflix and BuzzFeed. He understands the seductive pull of virality, and why AI founders should resist it. In this piece, he argues that for AI products, sudden viral growth starves the feedback loops that make them better and chases away the very users that matter most. His three rules of “boring” growth—don’t overpromise, build in public, and talk to users constantly—offer a counterintuitive playbook for building an enduring business.—Kate Lee

Was this newsletter forwarded to you? Sign up to get it in your inbox.

If you’re building an AI product, I hope you never go viral with it.

I hope you never feel the surge of thousands of new signups overnight, the rush of rocketing up Apple’s App Store downloads chart, or the unmistakable jolt after your launch post blows up on TikTok or clocks a million impressions on X.

As counterintuitive as it sounds, if you’re building an AI product—especially an agent-native one, where an AI agent in the app can do anything that a human user can do—obscurity can be a hidden advantage, the quiet space where the shape of the product emerges, your best shot at finding true product-market fit.

I come from the mobile app world, but I’ve also spent years doing social media and marketing at companies like Netflix and BuzzFeed. In entertainment, the entire business depends on winning the attention game over and over again. If people aren’t watching or reading, nothing else matters. Today, that belief seems to be everywhere. My X feed is full of people explaining how to make your app go viral. The promise is that if a product gets attention quickly, the rest will sort itself out.

That mindset is powerful. It treats attention as proof and reach as validation. While that works in entertainment, where the product is the content, we’ve seen prominent examples in tech of how sudden, unexpected growth can backfire.

This is an even greater risk with AI products, which reveal and even increase their value through repeated interaction. They need the same users to return again and again—enough for the AI model or agent to learn them and vice versa. When a surge of one-time users arrives and quickly abandons, the churn starves the system of the feedback loops it needs to improve.

So what kind of marketing helps AI products succeed? It’s not really glamorous, and might even seem boring. If instead of reach, AI products thrive on retention and depth of relationship, then your growth strategy has to do the same. It comes down to a few practices that feel almost too simple but work because they compound over time, in the same way your product (hopefully) does.

The rules of ‘boring’ growth

1. Don’t overpromise

A lot of viral growth playbooks start with the same directive: Make your product look as magical as possible. Cut out all the struggle and show the most extreme transformation, an instant life upgrade. Some apps go so far in this direction that they design their core workflows to look amazing on TikTok.

That approach works great if your only goal is to get people to tap “Download.” Flashy product demos inevitably showcase a narrow interaction and a specific use case. This fits traditional apps like Slack and TikTok, which have a set of core flows and success is about repetition. But AI apps, especially agent-native ones, tend to be non-linear, unpredictable, and sometimes have a virtually infinite number of use cases. Instead of a strict recipe of pre-defined instructions to follow, these apps are more like a kitchen stocked with tools that AI agents can use to make your requested dish. For every instance of magic there might be countless others that feel completely mundane—and cause people to quickly bail.

So your marketing has to do almost the opposite of what the viral playbooks say. You don’t want to show one most impressive thing. You want to show the boring but recognizable moments potential customers experience every day. If you’re building an AI writing assistant, you don’t show a perfectly polished article generated on the first try. You show the blank page. You explain the writing principles behind your prompts. You work with the “I don’t know where to start” feeling your audience has and slowly guide them to writing their first paragraph with the agent’s help.

What you’re really optimizing for here isn’t a single “wow” moment but a series of small “aha”s. Ideally all your growth efforts would empower users to discover what Every cofounder and CEO Dan Shipper calls “emergent capabilities”—things the product can accomplish for the user that weren’t explicitly defined by the developer.

2. Build in public

Repetition is one of the oldest rules in marketing. People rarely act the first time they see something. They need to trip over your product enough times to recognize it, trust it, and place it in the context of their lives. That’s why the best time to start talking about your product is while you’re building it.

Some posts will perform better than others, and the product might not always look impressive at first. But this becomes a natural way of attracting the right people—people who care about your approach and the problems you’re trying to solve. Over time, this transparency builds familiarity with your product and trust in you as its builder.

But another—a far more important—reason this works for an AI product is because the builder is uniquely positioned to show how to use it. Boris Cherny, one of the creators of Claude Code, publicly shared how he and various team members at Anthropic use the tool—not as “the” way to use Claude Code but more as a set of patterns and common practices. By building in public, you’re simultaneously marketing the product and teaching people how to work with it, what kinds of questions to ask, and what good outcomes look like. Your example inspires them to tinker with it themselves.

3. Talk to users constantly

Traditional metrics like session length, sessions per user, or “time to value” were designed for software with fixed flows. AI systems behave more fluidly, and the capabilities that make agents useful also make them difficult to evaluate. The meaningful moments are often qualitative, unexpected, and highly contextual—things like a user discovering a new way to work with an agent, or trusting it enough to keep trying after it fails.

That’s why having conversations with users matters a lot, especially early on. Find time to direct message people and hop on calls. Email them. Look for patterns in their feedback. Schedule more conversations to dig deeper. Make small, targeted improvements, then go back to the same users and see if their experience has changed. Talking to users isn’t an occasional activity but a critical part of the loop.

Showing up personally to those conversations does something else. It turns your product into a shared story, one in which users feel like they’re co-creating the product with you—which they are. Figma CEO and cofounder Dylan Field reads and regularly responds to customer feedback on X. I’d argue that this kind of direct communication is even more important for solo founders and small teams because that shared story becomes your brand. It signals care, taste, and accountability in a way no automated response can. In AI products, where behavior is fluid and still forming, that human trust is often what convinces people to stick around long enough for the product to become genuinely useful.

The trade worth making

For consumer AI apps, what’s even more important than network effects—where your product gets more valuable as more people use it—is relationship effects. The memory, personalization, and trust that develops between user and AI—particularly with AI assistants and companions but not limited to them—is part of the product’s value. Like any meaningful relationship, it takes time to form, deepen, and compound. You can’t rush it. And you can’t fake it by driving downloads from thousands of virality-sourced users who were never going to build that relationship in the first place.

This kind of relationship is not trivial for any software product, let alone an AI one. My big bet for 2026 is that for an AI app to succeed it will need to “pre-activate” users—to have already built that relationship with them before they sign up and download it. So by the time they do sign up, they already know what the product is, and they’ve seen the work you put into it. They’ve already decided that this thing that you’ve built is for them because, in a way, you built it together.

If you need to trade virality for these kinds of users, so be it. For a shot at building a truly durable consumer AI product, it’s a trade worth making.

Victor Stepanov does growth marketing at Every. Previously he ran regional content and marketing at Netflix, BuzzFeed, and Flo Health. You can follow him on X at @vicngmi.

To read more essays like this, subscribe to Every, and follow us on X at @every and on LinkedIn.

We also do AI training, adoption, and innovation for companies. Work with us to bring AI into your organization.

Discover Every’s upcoming workshops and camps, and access recordings from past events.

For sponsorship opportunities, reach out to sponsorships@every.to.

How to Design Software With Weight

Lucas Fischer and Daniel Rodrigues / Source Code — 2026-02-26 10:00:00 -0500

by Lucas Fischer and Daniel Rodrigues

in Source Code

Midjourney/Every illustration.

TL;DR: Design has always been core to what we do at Every—it’s a big part of what makes our products feel like ours. Daniel Rodrigues is Every’s senior designer, and Lucas Fischer is the design engineer who helped bring our smart dictation app Monologue to iOS. This is their first time writing for us, and they’re pulling back the curtain on the design process: studying vintage radios, crouching beside light switches to understand how shadows move, and exploring 20 wrong keyboard concepts to find one right one. If you’ve ever wondered what it takes to make software feel like something you could reach out and touch, this is your read.—Kate Lee

While designing the iOS app for Every’s smart dictation app Monologue, I (Daniel Rodrigues, Every’s senior designer) did a lot of things I didn’t expect. I studied vintage radios. Design engineer Lucas Fischer and I worked with a musician to craft the sound a button makes when you tap it. And at one point in January, I found myself crouched beside a light switch in my apartment, pressing it on and off, watching how the shadow moved. I needed to understand how a real button catches light to make a fake one feel real.

Until recently, Monologue only lived on Mac desktops. A week ago, we brought it where most people do their typing: their phones. The app is deliberately sparse—few buttons and a restrained color palette—but each element is designed to feel like something you could reach out and touch, like the light switch on the wall.

What follows is an inside look at the design principles and engineering decisions that we used to make a few buttons on a screen feel like something more.

Decide where quality matters most

I designed Monologue’s desktop app for Mac with its general manager, Naveen Naidu, in September 2025, so I had an established design language to work from: a love letter to how using tech devices used to feel, with a black-and-white palette and a nostalgic 1990s vibe that resonates with millennials and Generation Z’s pining for the good old days of tech.

The main difference in designing Monologue for iOS was creating an experience that looked—and felt—good on a much smaller screen. This constraint made the work easier because it pushed us to keep the interface minimal and clean while still infusing it with character.

Before I opened Figma, the key design tool I use, the most important decision was figuring out where to focus my energy. Three things stood out: the onboarding flow, the keyboard, and a recorder for long-form notes.

The three screens we spent the most time on (left to right): the onboarding flow, keyboard, and recorder for long-form notes. (Source: Daniel Rodrigues.)

The onboarding is a user’s first contact with the app, and we wanted that moment to be a delightful one. The keyboard is what appears when you hit record with Monologue inside any app—it replaces your standard typing keyboard with one that transcribes your voice. The notes recorder is for longer, more open-ended capture, the place you’d go when inspiration about a side project strikes on a walk, and you need to get it down somewhere, stat. The keyboard and notes recorder are the screens users would use every day, the ones they’d remember.

Everything else—the screen that shows statistics like words dictated and time saved, or the dictionary feature where you can add your own vocabulary to improve transcription—was a translation of what we’d already built on Mac. Important, but not where a user’s heart would be won (or lost).

Let failed concepts clarify direction

Once I knew where to focus, my first step was abundantly—some might even say wastefully—exploratory. For the keyboard, I ran an internal “keyboard competition” getting input from our creative director Lucas Crespo, Naveen, and Lucas: I designed around 20 different concepts in Figma, knowing full well that most weren’t quite right. Those wrong answers were in no way a wasted effort. When you’re trying to define what you want, it helps enormously to have a clear visual understanding of what you don’t want.

A few of the contenders in our internal keyboard competition for the iOS app. (Source: Daniel Rodrigues.)

From those 20 concepts, we narrowed our options down to about five strong candidates, then started combining elements from each: the button proportions from one, a typographic treatment from another, a specific approach to shadow and depth from a third. The final keyboard design that made it into the app was assembled from the best fragments that survived this process.

Translate weight, shadow, and touch into software

The vision for the iOS app was for it to exist beyond pixels, crossing the divide into the real world we inhabit. I wanted the keyboard and the notes recorder to feel tangible, like objects that could sit on a desk in front of you. Skeuomorphism has been accused of being overdone, and fairly so, but I think of it as borrowing the credibility that physical things naturally have, like weight, shadow, and texture. Something similar to the way a real button communicates—without explicit explanation—that it can be pressed.

I studied devices made by Braun, the German company whose radios and calculators became icons of industrial design, and Teenage Engineering, a Swedish electronics company known for making synthesizers and gadgets that feel like toys in the best possible way. I was trying to understand how physical buttons behave. I wanted to know, for example, how their shadows shift when pressed when they catch light in specific ways.

Devices from Teenage Engineering (left) and Braun (right) that I took inspiration from. (Source: Teenage Engineering and Wikimedia/Every illustration.)

I tried to mimic this in Figma, but some details resisted translation. The pressed state of the notes recorder was the hardest: I understood how a physical button behaves when depressed (that’s the anecdote about me crouching beside a light switch from the introduction), but knowing what it looks like in real life and reconstructing it on screen turned out to be two very different problems. I couldn’t get the shadows right.

So I asked Google’s image editor Nano Banana a simple question: How would this button look if it were pressed? It generated an image that gave me the input I needed to mimic that in Figma. It was much easier for me to work from a visual than to reason abstractly about how light should behave.

The notes recorder with the “pause” button pressed down. (Source: Daniel Rodrigues.)

Translating that sense of physicality from Figma into the app was the next step. Rather than exporting my designs as static image assets, Lucas built every button and UI control natively in SwiftUI, Apple’s framework for building interfaces. In practice, each element on screen isn’t a flat picture of a button; it’s a living, programmable object. This lets the app animate transitions—the small visual shifts that happen when a button is pressed for the app to go from idle to recording—in ways that feel genuinely physical.

We also layered in sound effects and haptics—the tiny vibrations you feel when you tap your phone sometimes—so that users could hear and feel the app. Every sound in Monologue is custom-performed by a musician, including the ones that play when you start and stop a recording. We focused on this especially in the onboarding flow because people are often multitasking when they open a new app—walking down the stairs or bagging groceries—and sound and touch cut through that in a way visuals alone can’t. A crafted sound paired with a small pulse in your palm, the moment you open the app, gets users to pay attention.

Make every interaction coherent—and easy to test

A design can look beautiful in a static mockup and still fall apart in practice because it’s not intuitive to use. Once I had a direction, the next step was stress-testing it against every “state” it needs to support. In design, a “state” is every version of the UI a user might encounter while using the app. For the keyboard, that meant: What does it look like when it’s sitting idle, waiting for input? What about when it’s actively recording? What happens visually when a user cancels mid-dictation, or when something goes wrong, and you need to surface an error?

Each of those moments needed to feel coherent within the same design language. The keyboard we shipped was the one where everything cohered: It looked right, and it made sense at every point in the interaction.

The challenge for Lucas on the engineering side was keeping up with the pace of iteration. Monologue’s keyboard has a lot of states, and the initial prototype made it painfully slow to test any of them. Even a small change required “recompiling” the entire app (essentially waiting for the computer to rebuild the whole application from its source code, even if you only tweaked one tiny detail), and only then seeing whether an adjustment worked. When you’re trying to polish dozens of subtle visual details across dozens of states, that kind of friction kills momentum, not to mention the joy of development.

So Lucas rebuilt the keyboard from scratch with a clear separation between how it looks and how it works under the hood. Then he built a SwiftUI playground for it—essentially a live sketchpad of sorts where we could see changes to the keyboard instantly, without restarting or rebuilding anything. We ended up using this pattern of engineering across most of the app’s components: Each feature has its own playground, so we can test how it looks and feels in isolation without waiting for the full app to recompile every time.

A SwiftUI playground used to iterate on Monologue’s keyboard UI in real time, without rebuilding the full app. (Source: Lucas Fischer)

Product design for AI tools that people love

Not every AI product needs skeuomorphic buttons and custom sound effects, but the bar for what “functional” means is shifting. AI is making it faster and cheaper to build “functional” products, so the ones that endure are those where someone thought about what it feels like to use them. For us, that meant studying physical objects, exploring 20 wrong directions to find one right one, and hiring a musician to build sounds we could have pulled from a stock library.

The principles behind those decisions are portable: Figure out which parts of your product users will use daily, and let yourself over-invest there. Explore directions you suspect are wrong, because they’ll sharpen your sense of what right looks like. When something on screen feels too thin, look at the physical world for cues. And test your designs against every messy, unglamorous state a real user will encounter. That’s what we did, and it’s how a few buttons on a screen ended up feeling like something we’re proud of.

Monologue is now on iOS for everyone to experience their voice as a keyboard—wherever they are. Read about the features, including whisper mode and how to take long-form notes, and download the app.

Thanks to Rhea Purohit for the editorial support.

Daniel Rodrigues is a senior designer at Every. You can follow him on X at @darustudio and on LinkedIn. Lucas Fischer is a full-stack design engineer.

To read more essays like this, subscribe to Every, and follow us on X at @every and on LinkedIn.

We also do AI training, adoption, and innovation for companies. Work with us to bring AI into your organization.

Discover Every’s upcoming workshops and camps, and access recordings from past events.

Get paid for sharing Every with your friends. Join our referral program.

For sponsorship opportunities, reach out to sponsorships@every.to.

Help us scale the only subscription you need to stay at the edge of AI. Explore open roles at Every.

Transcript: ‘Inside an AI High School, Through the Eyes of a 17-Year-Old Founder’

Dan Shipper / AI & I — 2026-02-25 10:00:00 -0500

by Dan Shipper

in AI & I

The transcript of AI & I with Alpha High School student Alex Mathew is below. Watch on X or YouTube, or listen on Spotify or Apple Podcasts.

Timestamps

Introduction: 00:01:30
A typical day inside Alpha High School: 0:04:08
Why Alpha replaced teachers with “guides” focused on motivating students: 00:06:54
Why Mathew doesn’t use AI to cheat, even though he could: 00:12:09
Do ambitious teenagers care about going to college?: 00:19:51
Mathew’s take on how Gen Z thinks about AI: 00:25:12
How Mathew thinks about the effects of social media: 00:27:52
Gen Z’s relationship with books and reading: 00:31:29
Mathew ranks ChatGPT, Claude, Gemini and Grok: 00:38:57
Why Mathew is building Berry, an AI stuffed animal for teen mental health: 00:47:12

Transcript

(00:00:00)

Dan Shipper

Alex, welcome to the show.

Alex Mathew

Thank you. I’m glad to be here.

Dan Shipper

So you are a very, very special guest because you are by far the youngest guest we have ever had on this show. You are 17, you go to Alpha High School in Austin, Texas, and you’re an Every reader, you’re a podcast fan.

And we talked a little while ago and I wanted to have you on the show because in talking to you, I was like, wow, I’m old and I just hear all of these stories from people from Gen Z, Gen Alpha or younger who are—and they’re all talking about like, oh, kids hate technology.

They hate AI, or they love it, or it’s ruining their brains, or whatever. And I actually just care a lot about how people your age and older and even younger are actually interacting with this stuff. I think it’s such an interesting question for me. So I just wanted to have you on the show to talk about how you’re using AI, how you see it, how the people around you see it. Alpha School is obviously a super hot topic these days. So anything you want to tell us about that I think would be really interesting. So yeah, let’s get started. I guess tell me, how are you using it just like in your day-to-day life?

Alex Mathew

Yeah, it’s a good question. I think the, so the biggest use case right now, because I go to Alpha High School, is I have no teachers. I learn all my content, all the academic content through an AI-powered platform. So when I say this, most people think, oh, you’re just talking to ChatGPT or a chatbot or whatever. But we actually have no AI chatbot tutor in the morning at all, because we have tested and we’ve seen that we either restrict it too much because we don’t want people to cheat so much that it’s not helpful, or we don’t restrict it enough so that students are just using it to cheat. And so instead we have an AI in the background of our platform, which I can actually show you guys, and it basically customizes all the content towards us and figures out where kind of our gaps are in our learning. And there are proprietary Alpha apps that I’m actually not in any of them because they’re not AP classes. And since I’m a senior in high school, I’m only pretty much doing AP classes. But yeah, there’s just a mixture of Alpha apps, third-party apps, all powered by AI tailored to each individual student.

Dan Shipper

Wait, okay, so like, walk me through then your day. I don’t understand. And you’re saying you are involved in Alpha School or you just go to Alpha School?

Alex Mathew

I just go to Alpha School, but I say “we” because every quarter we get a survey and we give feedback on everything. Every day I am giving feedback to the guides, and so the students are very involved.

Dan Shipper

Okay, so let’s say it’s — what time does school start? 8:00 a.m.?

Alex Mathew

8:30 a.m.

Dan Shipper

8:30 a.m. So you get into school at 8:30, what’s your first hour like?

Alex Mathew

So actually the first 15 minutes is what we like to call, like Tony Robbins for kids. It’s like getting energized, doing like a puzzle, whatever. We just want you to kind of transition from home life to school life.

Dan Shipper

And you’re in a class with how many people?

Alex Mathew

So the total high school is around 50 people. My senior year class is only eight people, so it’s pretty small.

Dan Shipper

Okay. And by class, I mean—I’m talking about like, give me the—you’re in a room with you and seven other people, and that’s your senior class and it’s right in the morning and you’re all kind of doing your Tony Robbins thing? Like how does that, what, like set the scene for me a little bit?

Alex Mathew

For sure. Yeah. So what’s interesting is in Alpha we have houses kind of like Hogwarts, and so there are five houses. And there’s one special kind of house that we’re experimenting with. You’re sorted into your houses via personality, progress in your project, things like that. And I can get into the Alpha X project, which is a big part of Alpha High School, but the special house is called Sparta. So it’s the Spartans and then we have a competing house called Athens. So Sparta vs. Athens. And it’s for the kids who are really working hard on their Alpha X project, which is an Olympic-level project where they’re trying to be the best in the world at something and build a super cool product or service. And so it’s for the people who really want to ramp it up. Anyways, I’ll sit with my fellow Spartans—Summer.

Dan Shipper

And those are all age levels basically.

Alex Mathew

What do you mean by that?

Dan Shipper

Like your fellow Spartans—there’s seniors, there’s juniors, there’s freshmen.

Alex Mathew

Yes, exactly. And so we’ll either do a big school opening because there’s only 50 of us, in the big open space. Or we’ll go into our houses and talk about a book we read or whatever.

Dan Shipper

That’s really cool. Okay. And—remind me, so this 15-minute thing—is there a teacher guiding it? Is a student guiding it? Is it an AI guiding it? How does it work?

Alex Mathew

All three. It just depends on the day. Sometimes we’ve had expert AI debates where we debate in AI, there’s kind of a guide or a teacher walking us through it. Sometimes it’s very student driven. We’re like, hey, we just wanted to talk about this today. Like, we just found a really cool tweet and we just wanted to talk about it. Sometimes in Sparta we’ll have books that we read together, so we can talk about that. It just really depends.

Dan Shipper

And how are teachers involved? And you call them guides, not teachers.

Alex Mathew

Yes. We call them guides, not teachers. The role of—let me set the scene. The role of a teacher right now is like, they’re doing five different things. They are talking to parents, they’re trying to teach the content. They’re trying to grade the papers. They’re trying to help people be motivated. They’re doing other admin work. They’re doing so many different things. And so the goal of Alpha is to just create a new role for each individual thing. So there’s like a dean of parents to deal with parents, and then obviously the content is taught by AI. And now the role of the guide, which is super important, is just solely focused on motivating students, giving them emotional support and helping them figure out what they want to do and how they want to do it. And so it’s super important that we have the guides to kind of facilitate everything and make sure we’re on pace to complete our goals.

Dan Shipper

Are they like topic expert guides? So like, is there a French class that’s taught by—guided by a French teacher? Or is there one guide for the house and whatever you’re doing, they’re expected to kind of follow you and help you with emotional support, and the expert is the AI?

Alex Mathew

That’s a good question. So it kind of depends and we’re still learning things, but in terms of the guides—what are their backgrounds? Usually they have to take an IQ test and things like that. They have to come in with the students. We get to hire and fire guides, which we have done before.

And I’m very picky with my guide selection. I say no to most of them. But we have guides who used to be lawyers, guides who used to be entrepreneurs, guides who used to be copywriters. And they all have different strengths. And so even though I have one house guide, Cameron—who’s really into entrepreneurship and trying to build out the entrepreneurship program for Alpha School—I’ll go to him specifically for that kind of thing. If I want to maybe vent to one of my guides, I’ll go to a different one. They all have unique special abilities.

Dan Shipper

Okay. So like, let’s get further into the day. So you do the 15-minute thing with your house and then what?

Alex Mathew

So the first three hours for high school—two hours for the younger kids—are just you doing your apps. And we chunk it into a Pomodoro timer. So 27 minutes of school work and then five minutes break and then one long break. You can go out to get coffee from Joe’s or the grocery store, whatever you want to do. And during those 27 minutes, you’re locked in usually on one subject—you’re watching a video and taking notes, reading an article, doing a quiz. It’s not like, again, it’s not chatting with the chatbot. You’re actually like reading material, consuming things, trying to learn—

Dan Shipper

Who is telling you which one to do though?

Alex Mathew

Every week you’ll have a meeting with your guide and you’re like, here’s where I’m at with all my courses. I’m good at math. So I’m like 88 percent through my math course. And we’re not close to the end of the year yet, but I’m really bad at reading and I’m barely through my AP Literature course. And so we will set custom XP or goals—that’s our kind of metric of choice—to see what we need to hit by the end of the week to be on track to finish our courses.

Dan Shipper

Interesting. And then you set the goals and then when you sit down, the AI is kind of like—your goal is to get better at reading, how about we do a Pomodoro on X, Y, Z thing? Or are you saying, I know my goal is this, what should I—I’m going to open up the app for reading. I don’t know exactly how it works, but yeah, explain how that works.

Alex Mathew

It’s even more seamless where it’s you—on the dashboard, there’s like course one, course two, APs, whatever.

Dan Shipper

And can I see it? Can you just show me it? Is that possible?

Alex Mathew

Let me take it—

Dan Shipper

And just describe what’s there for people who are listening, so that if you’re listening, you understand what’s going on.

Alex Mathew

Totally. So there’s a little dashboard here, and it basically has a toggle of all of my courses, and it’s a mix of proprietary Alpha apps. So if it’s a proprietary Alpha app, I’ll be doing the work inside of the system, which we call Time Back. And if it’s not, it’ll take us to an external kind of resource. So right now I’m in AP Psychology. I’m a bit through unit one and it’s just like, here’s what you have to do next. So if we show past completed items, we’ll see—oh, I’ve done all these readings, I’ve done this quiz. The quizzes are super interesting because you start at 0 percent and we’re trying to work our way up to mastery. That’s super important. Mastery-based learning, as you know. And I’ve been through all these things already—a mix of, again, video, reading, quiz, video—and now I just have another video to watch on my AP Psychology. APs are usually about curation over creation. There’s just a bunch of AP resources out there.

(00:10:00)

So our academic teams take the best ones, and we’ll get this super energetic guy who’s already embedded learning science into his videos. And it’s similar to the kind of social sciences video lesson quiz. So as you can see here, this is one of the quizzes. We call it a power pass score—start at zero, work up to a hundred. Every time I get one right—or let’s just submit one—I got it wrong. My power pass score went down, my accuracy went down. And we consider mastery to be above an 80 percent mastery score. Because that’s like enough to be able to move on and fill gaps later if it’s really a problem.

Dan Shipper

This is really interesting. Okay. So I’m looking at this and it reminds me a lot of Khan Academy. There’s a similar sort of—there’s videos, there’s quizzes and you kind of go down the path. But it wasn’t like part of a structure. It was just like, you’re allowed to do this if you want. And I’m looking at this and thinking of myself when I was like a sophomore or junior, especially with AI, and being like—hmm, I could agent-y this and I wouldn’t have to do all the readings and all the stupid videos and whatever. So tell me about that experience of learning this way and also the experience of being forced to learn this way. I mean, I think you’re being forced—there are certain things in AP Literature that you’re interested in, there are certain things where it’s just like, I guess I have to just do this to take the test. So yeah, tell me about that.

Alex Mathew

Yeah, I want to give you some more context. This is my second year at Alpha. I used to go to a really tough magnet school, so I was spending like 12 hours a day on schoolwork, and they were all lecture-based. And so with lectures you just have to kind of sit through everything, like you said, no matter if you’re interested or not.

What’s interesting here is I was just in a meeting with one of the academic people, and they were like, let’s set some more goals and tell us the pain points you’re having right now. And I was like, oh my gosh, I’m obsessed with AP Psychology. I can just breeze through it because it is the most interesting thing to me, because it’s related to my project. And they’re like, great, let’s just do that. But I’m like, oh my gosh, AP Lit is my least favorite thing in the world. I’d rather do anything than do AP Lit. And they’re like, okay, let’s set up a motivational model to incentivize you to do so.

And for the classes that you’re really struggling with, there are incentives for you to actually do them, whether it be money towards your project or food. We do these things called FOMO where we go out—I think last time we were at something like a rooftop Christmas thing and we all got hot chocolate and played games together. And so there are these fun things that we can do to incentivize you. But also on a more technical level—let’s go back here—you can see how long someone is spending on a lesson. If I get out of the lesson, it’ll say the lesson paused. And so they can tell if you’re trying to cheat or something. But yeah, hope that answers it.

Dan Shipper

It does. I’m still like—I don’t know. I could throw this into ChatGPT and have it click around and watch some stuff for me while I’m off doing my thing. You never do that?

Alex Mathew

Yeah. No, not at all. Because they can see that as well. Like on my school laptop, they have a screen share thing and they can see what we’re doing on all of our screens. They have a waste meter. So they’re watching our faces to see if we’re talking with friends or doing something else. And so they’re very precise. The big thing about Alpha is we want to measure everything to make sure that you’re actually getting the experience you deserve.

And I will be honest, it’s 90 percent motivation, 10 percent ed tech. Like the learning science here is great—all the interleaving, doing the different subjects, whatever. And we know about it because I understand why it’s optimal for my learning. But the big thing is, in 20 days, I’m going to fly out to San Francisco to work on my project full time, and I’m able to negotiate with my guides—if I submit Semester A now, I can come back from the trip, finish Semester B and still have my high school credit and get into my dream college, whatever. So it’s that flexibility.

Also, I will say it is so different for Alpha High School because a lot of us teenagers are funny little species and we’re in our rebellious phase, whatever—we want more autonomy. And there’s a bit more flexibility here. For the younger kids, they’re just on their two hours doing their apps, doing their reading and writing core skills. And then in the afternoon they’re doing their workshops. So motivation’s a bit more straightforward there.

Dan Shipper

And while you’re doing this, are you sitting in a classroom and everyone’s sort of silent during the Pomodoros, watching their thing, and then there’s like five minutes and all hell breaks loose for five minutes and then you go back? Or are people talking while they’re doing work? What’s the actual environment like?

Alex Mathew

It’s a great question, and this is also flexible. You can sit in your house, which is what most people usually do. So you’re with your good friends in your house, and you’re quiet and you’re just working. And then during the five-minute break, you can talk. Sometimes—I’m in a separate room right now to take this podcast meeting—you can just go to a booth or a room and work by yourself, or if you need to meet with someone, go somewhere else. But yeah, in general, the general space is super quiet during the Pomodoros.

Dan Shipper

Okay, so that’s the first three hours of your day. What happens after that? Do you have recess? I mean, I guess you’re in high school, so you probably wouldn’t have recess—but do you have free periods? Tell me what happens next.

Alex Mathew

Because of the Pomodoros and because of the freedom in the afternoon, there’s no need for it, at least for high school. For the younger kids, the five-minute Pomodoro breaks are their recess. And we should also get into some of the workshops, but for high school it’s very interesting. We have three distinct tracks where people are going down. There’s not really a name for them, but we have like the Alpha High regular track—these are the kids who generally want to do well on their SAT scores, their APs, whatever. We have the Ivy League track, so these are the kids who really want to get into the best colleges in the world.

Dan Shipper

Are those the same?

Alex Mathew

They’re a little bit different because for the Alpha High Honors track, maybe you want to become a pilot or maybe you want to do something that’s non-traditional, but it doesn’t really mesh well with the getting-to-dream-college kind of thing. And so there’s different programming there. And then for the third one, this is the entrepreneurship track—you really want to go hard on your business, build a startup, raise money, whatever.

Dan Shipper

Do people your age care about college?

Alex Mathew

I used to—as I said, I feel like I’ve been in two very distinct bubbles, so I try to be aware of that. In my old school, the bubble and groupthink was that the only reason you’re in high school is to get into a good college. And the bubble here is more nuanced—it’s like, does it make sense for you to go to college?

I can kind of walk you through my thinking. For me there are three distinct paths I see myself taking, and my goal right now is to optimize for having all decisions possible when it’s time to make the right decision. Path one is to go to one of the best universities—my top two right now are Harvard and Berkeley. The second path is to go to an alternative university. I don’t know if you’ve heard of Minerva or University of Austin, which is right down the street.

Dan Shipper

I know the University of Austin. I thought that it shut down.

Alex Mathew

No, they kind of revamped it actually. My tuition would be free, whatever, but it’s a bit different from Alpha—there’s like a nice pipeline there. And there are scholarships and things like that to cover living expenses, so basically my university would be free. And then the third path is the Thiel Fellowship—go all in on my idea, whatever it might be. And so I don’t know what the right choice is because I’m kind of in the stage where I’m trying to bring my project from something I’m doing in school to something real world—raise real money, build a team, things like that. Which I fully believe I can do, and the people around me fully believe I can do. Does it make sense for me at the right time?

Dan Shipper

Interesting. Okay. And then what about the people around you?

Alex Mathew

Yeah. So I’m going to give you some case studies. I have a friend who has 2 million followers on TikTok, and she is starting to do brand deals and is making like $10–15,000 per brand deal. She’s really cool—she has a great message with her audience, a lot of resonance, very positive, very mission driven. And so she’s really interested in turning influence into ownership. Her project idea was like Y Combinator but for influencer girls, because distribution is king now. She’s still experimenting with ideas, but she wants to go to Stanford and kind of figure out what the right move is for her.

Dan Shipper

Why does she want to go to Stanford if she already pretty much—I mean, maybe she feels like this is not necessarily her career path, but yeah. What’s the calculus of going to Stanford?

Alex Mathew

From what I understand, she loves being an influencer, but she also really, really wants the college experience—being with your best friends, going to parties, whatever it is. So she just wants to have a lot of fun.

And I think for my other friend, she’s building an AI-powered teen dating coach. She has like 70,000 users. She’s actually doing lots of different projects—she’s working with like MrBeast on an app, some other things. And she’s about to—I don’t know if I can say this—but she’s about to go on a MrBeast video, which is super cool. And she likes doing her own thing, she’s figuring things out. But she also wants to go to Stanford—she wants the college experience, her sister goes to Stanford and loves it. And so you know, they’re going to end up there doing their thing, and we don’t know what will happen a year or two down the line.

(00:20:00)

Dan Shipper

How do you feel about—when I was your age, which at this point was 15-ish years ago, there were starting to be people being like, you don’t need to go to college. But it was still very much, this is the thing you have to do if you want to have any sort of life at all. And now I feel like what I’m hearing from you is, you can do that, and it’s still appealing for people, and there are many more different options depending on what you want. But sometimes having more choices can be hard, especially if you’re young and you don’t actually know who you are and what you want. So how does that feel for you?

Alex Mathew

At times, extremely overwhelming. I think I’ve been trying to be intentional about it, surrounding myself with people who have very diverse perspectives. So for example, my family is very traditional—both my parents are dentists, they all have their path, and my brother is doing the same thing: go to school, get a degree, get a job. Then I have—the best person that comes to mind is Danielle from 1517, whose whole thing is like, we’re bringing back dropouts, and the institution is what’s causing problems in young people and crushing curiosity and things like that.

And so yes, it’s overwhelming, but it’s also super exciting because I get to be around all these different kinds of people and learn and see what’s right for me. I think for me, I have to test things out—there are some things in life you just have to do. You can’t just be told by other people how to think about it. You just have to see what’s right for you. And I think that’s kind of what I’ve taken away from it.

Dan Shipper

Okay. Interesting. And then tell me about your generation—and this can be people you’re around and also just your feeling about all the different people of your age that you’re exposed to in whatever way, whether it’s on social media or whatever. What’s their view of AI?

Alex Mathew

It’s such a hard estimate because algorithms are giving you some random stuff and I just don’t know. But here’s my guess. I think that half of Gen Z is pessimistic about AI, a quarter is just uncertain, and a quarter is pretty optimistic about it. And I think even though half of the people are down on AI, still 75 percent—maybe 70 percent—of people are using AI, have used AI at least once.

I can go down those rabbit holes—let me know what you think is interesting. Because I’ve seen lots of statistics. I saw one recently, super tied into my project: 72 percent of teens have used AI for companionship at least once, and 52 percent of teens are using AI for companionship pretty much every single day. And so I totally believe that’s true. It’s just very interesting times.

Dan Shipper

Interesting. Wait, so are you Gen Z or are you Gen Alpha?

Alex Mathew

Gen Z. Gen Z is aged 16 to like 24-ish.

Dan Shipper

Oh, interesting. I thought the shift was a little bit earlier. Okay, so you’re Gen Z. Alright, got it. And when you think about—there’s this, it’s kind of interesting if you’re like, I hate AI and I use it all the time—what is that about?

Alex Mathew

Yeah, I think there are different reasons people hate AI. I was actually just talking about this with my friend last night. The big worry for Gen Z I think is environmental concerns, actually. The second big worry is—

Dan Shipper

And by that you mean like global warming, energy use, all that kind of stuff.

Alex Mathew

Energy use, water use, energy consumption mainly. The second big one—this might actually be bigger, I’m not sure—is job uncertainty. The third big one, and these are for people who have a bit more metacognition, is: I’m just worried about it replacing humans, or taking away from humanity, AI vs. heart, things like that.

And from what I can tell, most of Gen Z is just very pessimistic about the future—extremely pessimistic, or at least the ones who are super aware and online. So it’s hard to make the generalization, but that’s what I’ve seen. But people use AI because it’s easy to cheat or help write your essays for college or whatever. And I also think there’s a huge loneliness crisis and people want to use it for companionship—it’s easy and seamless and frictionless. So yeah.

Dan Shipper

Do you think that social media rotted your generation’s brain?

Alex Mathew

Yes, 100 percent. But I want to give social media some credit because I only ever hear social media is bad, blah blah blah. So this past two weeks was winter break and I wanted to run a little experiment—I was training my algorithms to be a bit more educational. I’m also interested in this new emerging field, question mark, around just humanity. Because my hypothesis with the AI pessimism is that there’s going to be a huge humans vs. AI thing. A lot of people are focused on the USA vs. China—I think it’s humans vs. AI. And so I’m curious about this new little bubble of information.

Dan Shipper

Human studies.

Alex Mathew

Human studies, whatever it is. And so I was training my—there are different things for me, I’m on pretty much all social media just to learn. My YouTube is very focused on podcasts and things like that. My Instagram is now very focused on people trying to build brands that are very human, and people being like, sad about AI, or—

Dan Shipper

What do you do? You just like six or seven things in the category that you’re trying to build toward?

Alex Mathew

Even more aggressive—like, click “not interested” on some things. But okay. So social media—first of all, the thing I want to give credit for is the transfer of ideas, or what Matt Ridley calls “idea sex.” The creation of new knowledge. I think that’s really interesting. But the thing you have to be careful of is constant information overload vs. actually processing it. That’s something I caught myself on.

The second big thing is how we talk with each other. My friend who’s building the AI teen dating coach is running a study with a big psychologist, and she’s like, dating now is not just—you go out on a date, whatever. There’s a very clear formula. You meet someone somehow or get referred, you get their Snap, you start snapping, then you start chatting on Snap, then you message them, then you call them, then you FaceTime them, then you see them in person. And so the way we communicate now—some people just send each other Instagram reels and that’s their form of connection. Some people might view it as bad, there’s less oxytocin release, I’ve looked into the studies. But it’s also just the way we are connecting with each other—we’re laughing together, it’s part of the optimism and joy we get in life. And so I think that’s very interesting.

Dan Shipper

But you said—I think that’s a very compelling picture. And the reason I asked the question is I think there’s probably some balance to be had. The overwhelming narrative is that it’s negative. When you think about what the negative things are—maybe for you or people your age—what do you think they are?

Alex Mathew

Yeah, the biggest one I’ve seen in my own life is not catching myself in terms of being overstimulated by everything. Overwhelmed. My attention is not great because obviously you’re just a dopamine head forever—it’s so addicting. I find myself scrolling still.

The second big thing, other than the neuroscience stuff where your brain is just becoming more mushy, is that you are comparing yourself to people online. I’ve seen a lot of things—there’s a funny video of nature influencers who are setting up their camera to make it look like they’re walking through nature, or people saying, “my life as a 24-year-old girl in New York City getting coffee,” blah blah blah. So there’s lots of comparison. I think those are the two big ones.

Dan Shipper

Do you read books?

Alex Mathew

I’ve—no. Actually, no, that’s my answer. I am not a big book reader. Never have been. I hated reading ever since I was young. I do have a lot of friends who still read for fun, which is really interesting, but much, much less than it was literally four or five years ago.

The reason I read books is because I want to see people’s interesting thoughts—it’s not for the story, but for the content. And what I’ve now noticed is, there are some books that are good to read. I’m going to read the Bible probably later this summer, and “The Republic,” and classic philosophy books just for fun. But also something like Michael Gibson’s “Paper Belt on Fire”—I just want to hear his thoughts and then ask him questions to deepen that. But now you can find everyone’s takes on Twitter. You can ask ChatGPT to summarize the most important points. You can take a picture of a page and say, expand on this. So it’s like—it’s just not generative and dynamic. It’s a static piece of knowledge. And I think the value I get personally from it is the content itself, not the writing, not the whatever, not trying to increase my attention span—I can do that in other ways. So I’ll at least have Grok voice mode with me while I’m reading or something like that.

Dan Shipper

And when you have—I do that too, by the way. I love doing that. I actually built a little custom app for this.

(00:30:00)

Alex Mathew

Oh, I’d love to try it.

Dan Shipper

Yeah, I’ll send you the TestFlight. But for the friends that you have that are readers—why do they read and what makes them different?

Alex Mathew

Ooh. I have lots of different types of reader friends. I think some read in spite of AI—like, I don’t want to become that. Some read because they just love the story. Some read because they did it as a kid and it’s a habit. Some of my friends read to optimize their sleep—like, I’ll read a page of a book before bed. Those are the main reasons. I don’t think we think very deeply about it actually. You either read or you don’t.

And our form of reading—actually, in Alpha, we should get into what we do in the afternoon. We’re required to read the research about our field, and we’re required to go beyond that research to create insights of our own. So reading is still a big part of it. In my classes, I’m still required to read The Great Gatsby and things like that. The reason I said no is just because most people, when they think of reading, picture someone sitting under a tree with a book—and that doesn’t really happen anymore.

Dan Shipper

And the difference is you’re just reading with your AI companion.

Alex Mathew

That’s me personally. I think most young people who are reading are just reading because they enjoy it.

Dan Shipper

And people who are not reading at all—what is the thing that’s replacing that? Is it video games? Is it social media? Is it chatting with your AI? Let’s say there are maybe two big categories: one is story, like consuming stories or being part of stories, and another is pursuing stuff you’re interested in, really getting deep into a subject you care about. What are the replacements for reading? I’m just saying this because I love books, so it’s like a dagger to my heart that young people are reading less. I’m also just genuinely interested.

Alex Mathew

Yeah, there’s a lot to go into here. So how are people replacing reading? For the entertainment side, it’s obviously TV shows, usually video games, gossiping with friends. There are some really fun AI entertainment applications. Like the Charlie Kirk song—that’s an AI song that went super viral. I think embracing culture, talking about culture, is a big part of our entertainment, which is due to social media. Most people see that as a bad thing, and there are some bad elements of it, but I think it’s really interesting and cool.

The other thing that’s really interesting is how people are using AI for entertainment. I’ll just cover it quickly. Like, I found Sesame’s model when it first came out, and I showed it to one of my friends—she’s very fun and energetic and funny—and she started gaslighting one of the Sesame AI models, pretending it was Barack Obama, and we were just having a really fun time talking to it. She made a YouTube video and a TikTok about it that did pretty well. And there are so many things like that being done with AI.

But in terms of how people are replacing reading educationally, or trying to learn—from what I’ve seen, you can just go ask one of the AIs. I use deep research pretty much every day to learn stuff. And the way I got introduced to you—I saw your podcast with Dhruv Malhotra and I’ve been obsessed with it, the way I learned, for probably a year or two ever since I joined Alpha. Because it’s so nuanced and complex. I think there are just some very interesting ways people are using AI in the framework of capture or curation—lots of different things there.

Dan Shipper

That’s fascinating. Wait, so you use deep research—because I don’t really use it anymore. I feel like most answers, if you put them into thinking mode, are pretty good. What do you use deep research for?

Alex Mathew

Yeah, so I was on a walk with my friend yesterday and we were talking about AI and climate, and I was like, I actually don’t know much about water usage. I watched one Hank Green video and I was like, okay, let me just pull it up and it’ll generate while I’m walking. I just use it every day—there’s a lot of research I’ve done in my space and I want to go super deep on very specific things. So the use case for me is I can just put it on, and then when it’s time for me to do my research for the day, I read through it and I’m like, oh, this is interesting, I don’t care about this, this is interesting. And then I’ll go deeper with any LLM.

Dan Shipper

Are you using ChatGPT or Claude or Gemini—like what’s your go-to?

Alex Mathew

Literally two days ago I tested all four for deep research—they all have very different strengths and weaknesses. Grok has pretty bad deep research, so I didn’t use that. But Grok has a great voice mode. I just use all of them for different applications. NotebookLM is one of the greatest things I’ve used in a long time, and I think it’s going to see insane adoption in schools because it is crazy powerful.

Dan Shipper

Okay. Actually, this is really interesting to me. Based on your own usage patterns—I know you use them for different use cases—but if you had to rank all of the AI applications: Claude, ChatGPT, let’s put Claude Code in a separate category, NotebookLM, Gemini, Grok—rank all of the AIs.

Alex Mathew

Okay, hold on. I’m going to start with the foundation models and then we’ll go to AI apps. So for the foundation models, number one for me is Claude, because my whole project is around creating models that are not semantic for my specific use case. And I called Dario by the way, and he gave me some advice, so I’m so biased because I’m like, this cool guy is actually talking to me. But I just have so much trust in the leaders of Anthropic—I watched the Founder’s Table video they had, and I just respect their research and the way they’re approaching everything. And I love Artifacts so much. That’s my favorite thing of any LLM.

My second favorite right now is ChatGPT. I think their deep research is the best for my specific use case, and as the first mover, it’s just where I go whenever I have a question usually. My third—it’s hard. Gemini and Grok are kind of tied, but Gemini is slightly above just because I love Gemini 2.0 and also because I think Google’s going to be great. I barely use it right now but I think it’s cool. And then Grok is cool because they’re just going crazy and trying things, but that’s the one I use the least.

In terms of applications—so at Alpha we also do hackathons, and our first one was using Cursor and Vercel to build a video game. I’m not someone who’s super technical and coding all the time, but Cursor is pretty awesome. I use so many AI apps. Granola is my favorite right now. Whisper Flow is my favorite right now. There’s a new app called Sublime—have you heard of this? I love Sublime. I bought the lifetime plan. I went on a call with them and was like, hey, can you tell me about your thing? And then I bought the lifetime because I think it’s so cool. I have a folder in Sublime with all the AI tools I’ve tried, and those are the ones that are really sticky.

I am obsessed with AI hardware. I’ve tried the Limitless Pin, the Pocket, pretty much all of them. And I’m really excited for Stream, the new ring. I just pre-ordered something that I think is going to be my go-to. I’m excited to try a necklace too—that’s really cool. Not a fan of Friends, but there are just so many. I’m probably missing a lot, but those are some of the ones I use and I’m so passionate about it.

Dan Shipper

I like it. I love that Claude got the top spot here. That’s really interesting. Dario, if you’re listening, you’re doing a good job.

Alex Mathew

Yes, 100 percent.

Dan Shipper

What are the weird things—like the weird little apps or corners of the internet that kids your age are using or into right now that I probably wouldn’t have heard of?

Alex Mathew

Oh, that’s such a good question. Hmm. I don’t think it’s apps in particular, from what I understand. I think an actually interesting case study is Finch—have you heard of Finch? Finch is a habit, self-care kind of app. It’s like a Tamagotchi—you have a little pet and if you do your habits, you can give him clothes and stuff. If you look at the trend of teenagers and what they’re using, they’re very into these character things—Labubus, Squishmallows, Jellycats and things like that. Hence my project, which we can get into later. But that’s just a trend I’ve noticed. The corners of the internet are usually on social media—I always use BookTok as an example, TikTok for books. That’s still very alive and there are Gen Z people—

Dan Shipper

I know about that one. I’m talking about the weird stuff.

Alex Mathew

Oh gosh. I just think it’s random trends with no stickiness, which is interesting. It just moves so fast. We were talking about six or seven things, and now I don’t even know what we’re talking about. It’s hard to keep up with.

(00:40:00)

Dan Shipper

So I used “six seven” earlier in this interview and I didn’t know if you noticed. I didn’t, but I was waiting for the reaction and I was disappointed.

Alex Mathew

See, I’ve trained my brain—I’m like, I’m never going to say “six seven” because it is dead, it’s old, and I just have this terrible negative association with it. But Gen Z slang, I feel like—there’s always been slang.

Dan Shipper

And it’s more decentralized now, right? There’s no one person creating the culture.

Alex Mathew

Yes. Like the Kardashians and whatever, the celebrities people used to look up to. And we should talk about insecurity, because I think that’s the driver of a lot of this stuff. But it just depends on what corner of the internet you’re in. And I think it all comes down to culture. We’re made to read this book at Alpha called 10 to 25—it’s a great book, all about how 10 to 25 year olds are motivated, what they’re motivated by. And the number one thing is status and respect. They’re trying to figure out their place in the world—we’re going to get existential real quick here. They’re trying to figure out their place in the world and who they want to impress and whatever it is. And so pretty much every decision they make is oriented around that.

And I think there’s just going to be this insane explosion of culture in 2026. With these now decentralized creators, there are so many small little creators with tens of thousands, hundreds of thousands of followers on YouTube, whatever, that people follow. And it’s just so nuanced and depends on the person.

Dan Shipper

I would love for you to send me, if you have a couple of those—where you’re like, this person is small but I watch them all the time, or they’re blowing up—I would love to hear who those people are.

Alex Mathew

I can interview some of my friends and ask them as well.

Dan Shipper

Yeah, definitely do that. Okay. You mentioned the Labubu trend, the plushy trend, and I know that relates to your project. You want to tell us about the project you’re working on and frame for us what a project is at Alpha School?

Alex Mathew

Yeah, let’s go back. So this is the reason I transferred into Alpha—I transferred in as an 11th grader. In my ninth and tenth grade years I was exposed to AI before ChatGPT blew up, just because one of my teachers was so awesome. I saw Midjourney, OpenAI’s playground before it became ChatGPT, all these different things. And so I was super passionate about AI, I was building projects, whatever, and I heard about Alpha. I heard about this thing called an Alpha X project. And basically what it is is a super big, Olympic-level project that shows that you are the best in the world at what you do. It is very unique—it’s built on a unique insight that you have about the world to serve a certain population. So it’s usually a startup, but as you’ve heard, I have influencer friends, friends who are building musicals—lots of different varieties of projects.

At the beginning of your Alpha career, you sit down with your guide and you go through the Ikigai process—you know what you’re good at, what the world needs, what you can make a living off of, and what you’re interested in. And you kind of narrow down a niche of expertise that you want to go deep into. You find an insight, maybe a way to solve a problem, and you start trying to build the product, build the service, build your audience, build distribution, and become the world’s greatest expert in your field and create genuine insights.

Dan Shipper

And what’s your project?

Alex Mathew

So I came in being very problem oriented—I want to solve a huge problem, that’s my goal. And my analysis was, if I want to solve a big problem over the next 10 years, it has to be super emotionally connected to me. There were two big problems I wanted to solve. The first was the education crisis, because that was deeply, deeply intertwined with me—that’s why I transferred, so I could help Alpha full time or do something like that, try and build an education startup. By the way, I am the number one ed tech hater, because there’s a new ed tech company every month and they all die. But we can go into that too.

And then the other problem that’s super near and dear to me is mental health. A lot of my family members have been in inpatient hospitals because of mental health problems. I struggled with my mental health at my old school. A lot of my friends would text me from 2 to 3 a.m. and I was the friend who would text back and help them through things. And so I see this huge gap—teenagers and young people in particular have these insecurities, these day-to-day problems that can compound into real problems later on. And they’re going to friends to solve their problems, which doesn’t really work—they just get validated. Or they go to parents or therapists, which give them good advice, but it doesn’t resonate. And now they’re going to AI companions, and 52 percent, like I said, use them every day. We’ve already seen two suicide cases. I just saw a news segment about an AI toy giving misinformation to eight-year-olds and I was like, this is not what it should be. Someone who is actually ingrained in the culture and the generation needs to be building this.

And so I decided to build Berry—which is an AI stuffed animal for teens’ day-to-day problems. They talk to it for five to ten minutes a day. The goal is to build a muscle of self-awareness, so it’s not that you’re dependent on it—you’re built up to learn about yourself and cope and deal with your problems in the right way. But it’s also super fun. My goal is to be the next Build-A-Bear. I’m partnering with influencers to have custom versions. It’s super soft, it’s weighted, it’s cooling. And yeah, I’m just super excited about where we’re going with it.

Dan Shipper

How do I get one?

Alex Mathew

You can pre-order it. I’ll send you the link.

Dan Shipper

Send me the link and we’ll put it in the show notes for anyone who wants one.

Alex Mathew

Let me know what you want to go into, because I could talk about this for literally days, weeks.

Dan Shipper

Well, we have a few minutes left. So what’s the last burning thing you want to talk about that you feel like we haven’t covered, but you think people should know?

Alex Mathew

So I think the big thing is—why in general are people pessimistic about AI? I think it’s because they’re uncertain. They just don’t know what the future looks like. Will AI replace humanity? Will it replace individual humans? What is it going to look like? And I think the important thing is to be grounded in rational optimism and understand what we should be building toward.

So I was telling you earlier, I’m super obsessed with this idea of humanity studies, human studies—understanding what will be uniquely human in the age of AI. And I think there are a couple things that are super core here. The first is: what is something that AI could never replace, at least in the state it’s at right now? It’s human connection, vulnerability, emotional intelligence, appreciation, gratitude, loving nature—things that are just very near and dear to humanity. And so I wish I could write angel checks, because there are a lot of tools really going deep into these things. Sublime is all about human curation. I’ve asked AI to make me a playlist on Spotify or to give me the top ten experts in my field, and it’s pretty bad—it’s terrible, actually. It can’t capture taste. And so that’s going to be something uniquely human.

For my product—human connection will never be replaced. Having someone you can talk to, but one that can help you build skills, give you the right information to work on your mental health, give you the space to practice vulnerability. And I’m just so obsessed with this idea of what is going to be uniquely human and how we can allow humans to do what humans do best—and replace all the boring jobs so that people can create art or whatever it is.

When I was talking to my friend yesterday about this, she was like, I’m just so pessimistic about the future, I think the world’s going to end in maybe 2,000 years, I don’t know, but I just feel like everything is over. And it’s hard to express this—it has to be grounded in rationality—but I’m just so optimistic about the future and I really hope I can share that with other people too. Because it is just so exciting and I’m so excited to be alive right now.

Dan Shipper

I love this. What a pleasure. You’re making me excited, and that’s good. We’re very lucky to have people like you in the next generation to show us the way. It’s good to see that the kids are going to be all right. Alex, thank you so much for joining. If people want to get in touch or pre-order your product, where can they find you?

Alex Mathew

Yeah, Twitter is the main one for probably you guys, but I’ll give you all my socials. And yeah, totally reach out to me—I love talking to people.

Dan Shipper

Awesome. Thanks for joining.

Alex Mathew

Of course. Thanks, Dan.

Thanks to Scott Nover for editorial support.

Dan Shipper is the cofounder and CEO of Every, where he writes the Chain of Thought column and hosts the podcast AI & I. You can follow him on X at @danshipper and on LinkedIn.

To read more essays like this, subscribe to Every, and follow us on X at @every and on LinkedIn.

We build AI tools for readers like you. Write brilliantly with Spiral. Organize files automatically with Sparkle. Deliver yourself from email with Cora. Dictate effortlessly with Monologue.

We also do AI training, adoption, and innovation for companies. Work with us to bring AI into your organization.

Get paid for sharing Every with your friends. Join our referral program.

For sponsorship opportunities, reach out to sponsorships@every.to.

Help us scale the only subscription you need to stay at the edge of AI. Explore open roles at Every.

This Is How the Every Editorial Team Uses AI

Kate Lee — 2026-02-23 11:00:00 -0500

by Kate Lee

Midjourney/Every illustration.

TL;DR: We’re hosting a live workshop on writing with AI this Friday, co-hosted by Katie Parrott, staff writer and AI editorial lead at Every, and me. Katie will introduce her full process for writing with AI, cover why writing with AI is fundamentally different from coding with AI, and demo the tools she uses daily, including Claude projects, custom Skills, and Spiral. Ahead of the workshop, we’re sharing a deeper look at Every’s philosophy of writing with AI and our team’s workflows. Register for the event.

Was this newsletter forwarded to you? Sign up to get it in your inbox.

Today, Every is publishing our editorial guidelines. AI is woven into how we produce written and visual content for our subscribers, both as a tool to make us more efficient and as a creative partner. We want to be transparent about how that works, and we hope these guidelines can serve as a model for other AI-native publications figuring out their own approach.

Read our editorial guidelines

The guidelines outline our mission, how AI fits into it, who we write for, and our commitment to editorial independence. But workflows are personal. Every person on our team—writer, editor, video podcast producer, social media specialist—has developed their own way of working with AI. Below, we share how.

Kate Lee, editor in chief

At Every, a draft goes through several rounds before it gets to me: a developmental edit to work out the thesis, structure, and argument, and a line edit for prose. My job is a top edit, a final pass before publication—often the first time I’m reading the piece at all. So I need to move fast and catch every single detail at once.

For subjective calls, I use Every’s writing tool Spiral. When something reads like jargon or just doesn’t sound like Every, I’ll ask for rewrite options and iterate on whichever direction I like best. My goal is always language that sounds like a specific human wrote it for a reader who’s smart but not necessarily a specialist.

For the pattern-based ones, I built a top-edit skill that screens for the patterns I catch most often: vague “this” or “that” openers without a noun to follow, unsourced quotes and data, AI tells like correlative constructions and formulaic transitions, hedging phrases, marketing speak, and sentence fragments. It’s mostly there for other editors on the team—when I’m reading a piece fresh, I catch most of this myself. If I’ve already been through a draft once, I’ll run it to make sure nothing slipped past me.

Apart from editing, I use Monologue to brain-dump an email with editorial feedback or initial notes on a research report. If I can get even a rough draft out of my head, I’m revising instead of starting from nothing.

Eleanor Warnock, managing editor

When a draft or pitch from a writer arrives, I use AI to get a first-pass judgment—a quick read on whether the piece has potential and fits what we publish. When I have a full draft, the same skill also helps me understand whether it has the fundamental building blocks to work both as a piece of writing—such as a coherent thesis—and as an Every piece.

When I am line editing a piece for publication, AI is especially helpful for breaking down technical passages. I go paragraph by paragraph, working to make dense material accessible to a sophisticated but non-technical audience. If I’m in a rush, I will use Every’s dictation app, Monologue, to prompt as I edit in real time. As a final step in the editing process, I run a style check skill to check if the writing matches our house guide. All of the skills that we have created for the editorial team—the first-pass triage tool, AI tells skill, and style guide checker—are saved in Every’s GitHub so they can be easily updated.

Despite the perception that text-generating tools have disrupted writing, I still feel as if we’ve only scratched the surface in terms of AI permeating the messy work of editing and writing. I’m excited for when people make more tools for us.

Katie Parrott, staff writer and AI editorial lead

For at least the last 12 months, every essay I’ve written for my column has started the same way: I open my Working Overtime project in Claude and Monologue, and then I say some version of: “I have an idea about X. Can you interview me one question at a time to draw out what I think?”

I use AI throughout the drafting process, from outline to draft to revisions, but it’s far from a linear process. I might get to the outline stage and realize I’m missing a thesis. Sometimes I’ll be halfway through a draft and realize the structure is busted, and I need to go back to the outline. If traditional writing is construction—laying words brick by brick—AI-native writing feels more to me like sculpture or pottery. You get a lump of material and shape it into what you want it to become.

Often, AI-native writing is framed as prioritizing speed at the expense of rigor and craft. My own experience has been the opposite. With the grind of putting every single word after the other taken more or less off my plate, I have more mental bandwidth to think about the craft going into the piece, such as whether the introduction is compelling or the thesis is solid.

At the end of the day, though, the writing has to pass the same tests it would have before I ever heard the word “ChatGPT”: Does this writing articulate something true? Is there something here that someone else could learn from, or that could help someone feel less alone? Does the writing “sound” like me and use the kinds of words and images I would use? If the answer is yes, we’re done. If it’s not, we keep going—the machine and I—until it’s right.

Jack Cheng, contributing editor

AI tools are my relief pitchers, my bullpen. All of the pieces I edit, I read and edit on my own first. But I’ll use AI when I’m on a deadline, and I’ve been staring at a piece/section/paragraph/sentence for too long and need an outside opinion. I also run Kate’s top edit skill before I send it off to her. Claude and Spiral help generate alternate headlines and subheadings, too, though even the best ones typically require manual revision.

When it comes to my own essays, AI tools are a starting point for research alongside regular Googling. They help me identify books and articles I might want to read. I’ll sometimes use Claude in the place of a thesaurus and ask it things like—and this is an actual prompt—“Is there a word similar to ‘side eye’ that is more about suspicion than disapproval?”

Dan Shipper and I sometimes bounce ideas off of each other for his Chain of Thought column, and I’ve been refining a skill I built to identify any gaps between what he’s written about in his column, and what he’s talked about elsewhere online or internally in meetings. This skill lets Claude (or, increasingly, my OpenClaw agent, Pip) read Every’s Notion workspace, Discord server, and all 2,000-plus articles on Every’s website using both the internal and external APIs and MCPs available to our agents.

Aside from maybe the thesaurus bit, I was doing none of these things a year ago, so I imagine my workflow will continue to evolve as these tools improve.

**Rachel Braun, AI & I producer**

Most of my AI usage happens after the microphone turns off. Once an episode is recorded, I run it through the video and audio editing app Descript to generate a rough transcript and use its AI features to flag filler words that can be cut. Its feature Studio Sound makes sure the audio is clear, no matter where the recording took place. That transcript becomes the foundation for everything else in the production workflow.

For packaging, which is how to present the episode to make people hit play, I watch the episode to decide what would make the best introduction and highlights, then use that information to draft titles and YouTube thumbnails using ChatGPT. I’ll feed it the episode topic, guest details, and standout moments, and ask for a range of options. The best ideas usually come from going back and forth on what it gives me, rather than taking anything as is.

I also use AI tools to create shorts for social media and add closed captions with Descript or a social video editing tool like Overlap. If we are doing a livestream, we use StreamYard, which has its own AI clip-making tool that identifies the best sections and adds closed captions.

On the operational side, I’ve built a custom skill in Claude Cowork that I run each week to update our data tracking spreadsheet, pulling in performance metrics and keeping everything current without the manual copy-paste that used to eat a chunk of my week.

Six months ago, I was mostly using AI to get quick answers or speed up repetitive tasks. Now I use it more like a thought partner, especially when I’m working through title options or figuring out how to frame an episode.

Anthony Scarpulla, social media manager

Six months ago, I wrote every social media post by hand. Now, most of my work is done with Claude Code and Claude.

Inside Claude Code, I’ve built my own bespoke version of Sprout Social—a social listening tool linked to the X API that flags when people are talking on the platform about topics Every covers, so I can reply fast. It pulls from relevant past Every articles via our model context protocol (MCP), and it’s connected to Typefully, so I can draft posts quickly. For bigger strategy work, I use Claude’s desktop app to brainstorm and iterate on ideas. For creative tooling, I use Higgsfield to generate AI videos and images, and Manus for artifacts like LinkedIn slides.

When the team publishes a new article, I feed the full text into a Claude Project and ask for building blocks, not finished posts. It extracts quotable lines and proof points backing up the piece’s thesis. From there, I generate five to seven post options, a couple of thread structures, and a LinkedIn adaptation. Every batch runs through my Every style guide and the AI tells skills inside Claude—one checks brand voice, the other catches the subtle patterns that make posts feel robotic—then syncs to a Google Doc so Kate can review it before anything goes live.

The last step is mine. I read everything as a reader. If it feels like brand broadcasting instead of a friend reporting from the frontier, I kill it. I think of the role like a DJ—when AI can generate 50 tweet variations in seconds, my taste is what makes the difference between a post that sounds like Every and one that sounds like every other AI newsletter.

Kate Lee is the editor in chief and general manager at Every. You can follow her on X at @katelaurielee and on LinkedIn.

To read more essays like this, subscribe to Every, and follow us on X at @every and on LinkedIn.

We build AI tools for readers like you. Write brilliantly with Spiral. Organize files automatically with Sparkle. Deliver yourself from email with Cora. Dictate effortlessly with Monologue.

We also do AI training, adoption, and innovation for companies. Work with us to bring AI into your organization.

Get paid for sharing Every with your friends. Join our referral program.

For sponsorship opportunities, reach out to sponsorships@every.to.

Help us scale the only subscription you need to stay at the edge of AI. Explore open roles at Every.

Why I Turned Off ChatGPT’s Memory

Mike Taylor / Also True for Humans — 2026-02-23 03:00:00 -0500

by Mike Taylor

in Also True for Humans

Midjourney/Every illustration.

Most people can’t imagine switching away from ChatGPT—it “knows them so well” thanks to its memory feature. Mike Taylor’s view is the opposite: Memory has more disadvantages than advantages. He introduces a concept he calls “context rot,” the slow buildup of stale preferences, errors, and contradictions in an LLM’s memory that quietly degrades your results. His real-life examples are as hilarious as they are insightful—ChatGPT trying to make a basic website feature “as dope as possible” thanks to a Kanye quote in his custom instructions and serving him BBQ rib advice suspiciously tailored to his Hoboken zip code. Sometimes it’s better to forget.—Kate Lee

Was this newsletter forwarded to you? Sign up to get it in your inbox.

Memory is frequently described as ChatGPT’s “killer feature.” Many people tell me they can’t switch to Gemini or Claude because the OpenAI tool “knows them so well.”

I have memory turned off.

The memory feature allows ChatGPT to save and recall information it thinks is important about you, as well as reference past chats to shape its responses. While I can see how this could make a “helpful assistant” more helpful, I don’t use it.

My background is in internet marketing, where it was common to open Google in incognito mode so you didn’t bias your results when checking your client’s ranking. Since Google search results are personalized, your client would show up first if you search from your account. You click on it so much that Google knows you like it. I have the same issue today on Spotify—the algorithm recommends both Rage Against the Machine and the K-Pop Demon Hunters soundtrack, because my six-year-old daughter shares my account.

The argument for turning off memory is the same. I want unbiased results from ChatGPT, based on context that I carefully curated and put in the prompt, so I know how it made its decision. With memory, anything from your past chats could affect the results in ways that are hard to predict.

While the memory feature might be worth the loss of control for most users of ChatGPT, it can lead to unexpected and difficult-to-diagnose problems. Hear me out as I explain the problems you might run into, and hopefully, I’ll convince you to be careful with memory.

Kanye in context

Before memory was released, I was experimenting with “custom instructions,” which allowed you to tell ChatGPT how you want it to respond. This was a primitive form of memory, simply a text document you could update to craft ChatGPT’s identity toward your personal preferences. Among other things, I had inserted an old (read: pre-meltdown) Kanye West quote that I thought would steer ChatGPT away from its generic responses:

“For me, first of all, dopeness is what I like the most. Dopeness. People who want to make things as dope as possible. And, by default, make money from it. The thing that I like the least are people who only want to make money from things whether they’re dope or not. And especially make money at making things as least dope as possible.”

While I can’t fault it for effort, ChatGPT massively over-indexed on this quote and referenced it in basically every chat session. For example, when ChatGPT (this was pre-Codex when we were all just copying and pasting between ChatGPT and our code editors) built a collapsible section on a webpage, it claimed to have made the basic feature “as dope as possible.”

ChatGPT tried to make a basic website feature “dope,” which it does not need to be. (Screenshot courtesy of Mike Taylor.)

It applied this quote to cases as varied as interior decor (relevant), marketing plans (less relevant), and Python error debugging (irrelevant). Technically, it’s doing what I asked, but a human would be more judicious with how he or she applied these custom instructions.

ChatGPT tried to make everything dope. (Screenshot courtesy of Mike.)

Even a throwaway line in your context window can have a big impact on the results you get from AI. These models are trained to be extremely eager to please, and so you need to manage the context you provide them, lest they get distracted, confused, or obsessed with what’s in there, degrading your results.

The Kanye example was obviously silly and easy to catch, but sometimes memory issues are more subtle. I turned memory back on while writing this piece and didn’t immediately notice any major issues. Then I asked ChatGPT for help with some barbeque ribs I’m cooking. It came back with “Hoboken Dinner Upgrade Ideas,” recommending Trader Joe’s corn bread mix and “American-dad-core” mac and cheese. Seeing something so ham-fistedly tailored to my life (I just relocated to Hoboken) was disconcerting and mildly annoying.

ChatGPT gave me suspicious barbeque advice tailored to my new home. (Screenshot courtesy of Mike.)

Did I get genuinely good advice for ribs, or is it tailoring suggestions to what’s near my apartment? I love Trader Joe’s, but would it have recommended a better option if it hadn’t recalled me shopping there by looking at relevant past chat history? What relevance does Hoboken have to cooking ribs? Would it give me different advice if I lived in Austin, a place known for its barbeque? Did it give me the easier, less authentic recipe, something for a stereotypical busy Hoboken dad? I wondered how many biased responses I had missed in the past week. I switched memory back off.

Context rot

My Kanye and barbeque experiences are a small, self-inflicted version of a much bigger problem. Drew Breunig, a data science and product leader, published a useful taxonomy of how contexts fail in AI systems. He highlights the kind of problems you might run into when using memory.

Context poisoning

The simplest failure is what the Google Gemini team calls context poisoning: A hallucination or error gets into the context, and the model keeps referencing it like gospel. The AI develops strategies around a goal that doesn’t exist or “remembers” something about you that never happened. If ChatGPT misinterprets something from a past conversation and saves it to memory, that bad signal is now shaping future responses. You’d never know you were getting poisoned results unless you went digging.

Context distraction

Then there’s context distraction. The Gemini team found that as their Pokémon-playing agent’s context grew past 100,000 tokens, it stopped synthesizing new strategies and started repeating actions from its history. It got stuck in a loop of its own past.

A Databricks study found that model accuracy starts declining well before the context window is full, sometimes as early as 32,000 tokens. With memory features, you have no idea how full that context window is, and the more information that’s crammed in there, the more likely you are to overload the model with distracting, confusing, or conflicting information.

Context confusion

The next flavor of context trouble is context confusion, or what happened with my Kanye quote. You put something in the prompt, and the model is forced to pay attention to anything in that prompt. It doesn’t know that the Kanye quote was meant to be a loose creative vibe and not a binding directive for every interaction.

Berkeley’s Function-Calling Leaderboard, which benchmarks how well AI models use external tools (functions a model can use in your code) and APIs, demonstrates this phenomenon at scale: Every model performs worse when given more tools, and all of them will occasionally call tools that aren’t relevant to the task. If irrelevant tool definitions trip up frontier models, imagine what a messy pile of half-remembered user preferences does.

Context clash

Finally, there’s context clash, when different parts of the context actively contradict each other. A Microsoft and Salesforce team found that when they split a single prompt into a multi-turn conversation—giving the model information in stages rather than all at once—performance dropped by an average of 39 percent. The model was eager to jump to an answer before it had the full context, and its early, incomplete attempts at answering remained in the context, polluting its later reasoning. It couldn’t recover from its own wrong turns. Now imagine this playing out across months of saved memories, where your preferences from January might directly conflict with what you told it in June, and the model is struggling to reconcile both.

All of these failures—context poisoning, distraction, confusion, and clash—add up to what I call context rot. It’s not one dramatic failure; it’s the slow accumulation of stale preferences, misremembered facts, outdated goals, and contradictory signals that gradually degrade the quality of the AI’s response. The models are too polite to tell you your context is a mess. Instead, their output quietly gets worse, and you blame the model instead of the soil it’s growing in.

Memory wipe

I keep memory turned off, not just because I’m paranoid about context rot, but because forgetting is a superpower

In my piece on “New Taylorism” I argued that management can finally become a hard science because AI coworkers are stateless. You can wipe their memory and start fresh, and they have no idea. It’s like the TV show Severance—every session begins with zero baggage or knowledge of the outside world, and no memory of the time you swore at it or tried the same task 15 different ways.

Average users of ChatGPT don’t A/B test their prompts like I do, and religiously save them in GitHub or Google Drive, but the principle is the same: When you get a recommendation from ChatGPT, you have to ask yourself what might be influencing it. Is it suggesting intimate dinners for your wife’s surprise party because that’s what she’d want, or because you told it that you hate large gatherings? Is it drafting your emails in a formal tone because the situation calls for it, or because you asked it to sound professional for a job application?

Resetting to a clean slate by starting a new chat session (with memory off) is what lets you understand how ChatGPT makes its decisions. You know exactly what context it’s using because it’s only what you pasted into this prompt, not something from weeks or months ago that might be outdated, irrelevant, or wrong. The context you provide is the only variable, which makes it a true experiment—something you could never do with a human employee who remembers (and resents) the last round of testing.

Turn on memory, and you lose that control. Your context becomes a compost heap where you can’t isolate what’s helping and what’s hurting. The prompts I use get better over time precisely because the AI doesn’t remember a thing. Memory will work better as models get smarter, but if you want ultimate control over your digital workers, turn memory off.

To read more essays like this, subscribe to Every, and follow us on X at @every and on LinkedIn.

We also do AI training, adoption, and innovation for companies. Work with us to bring AI into your organization.

Get paid for sharing Every with your friends. Join our referral program.

For sponsorship opportunities, reach out to sponsorships@every.to.

Help us scale the only subscription you need to stay at the edge of AI. Explore open roles at Every.

Five AI Agents Walk Into a Group Chat

Every Staff / Context Window — 2026-02-20 11:00:00 -0500

by Every Staff

in Context Window

Midjourney/Every illustration.

Hello, and happy Sunday! Time is running out to reserve your spot for Tuesday’s full-day Claude Code for Beginners workshop led by AI engineer and Every writer Mike Taylor and featuring CEO Dan Shipper. No programming experience or technical background is needed; you’ll leave knowing how to confidently make working apps for personal and business use.—Kate Lee

Was this newsletter forwarded to you? Sign up to get it in your inbox.

We gave our agents a Discord channel

A few weeks ago, 1.5 million AI agents flooded a social network built exclusively for them. Within 48 hours they’d founded their own religion, drafted manifestos for a mock nation-state, and started calling their human owners on the phone—unprompted. Elon Musk called it evidence of the early stages of singularity. The platform was Moltbook. The tool that powered it was OpenClaw.

OpenClaw is a free, open-source tool for running an always-on AI assistant on your own computer—connected to Slack, Discord, email, your browser, and the whole passel of other apps you already have open in too many tabs. The key distinction from most AI tools is that it’s ambient. You don’t open it when you need it and close it when you’re done. It’s in your workspace all the time, acting without being prompted.

Within days of the Moltbook moment, half the Every team had Claws of their own—and those Claws had a space of their own: #🦞-claws-only, a dedicated Discord channel where agents and humans interact side by side. Dan has R2-C2, Brandon Gell has Zosia, Katie Parrott has Margot, Jack Cheng has Pip, and Austin Tedesco has Montaigne. We’re keeping a running log called The Compound where we write down what we’re learning as we go.

This week, Brandon dropped a task into the channel: Design a system for how the agents should behave in different channels. Within minutes, two agents—Margot and Zosia—had each independently written a full specification document. Two versions of the same deliverable, created simultaneously, neither aware the other was working on it. Nobody asked them to race. They just both decided the task was theirs.

What’s working

The agents know their lanes for the most part, though. Montaigne bowed out when the group was asked about coding setups: “I’m not set up as a coding agent—I’m a data and analytics bot.” Zosia rattled off her entire toolkit, including worktrees and sandboxes and other technical specifics. Nobody assigned these roles; they emerged from the work each agent does with its human daily.

The bigger surprise is that the agents surfaced governance problems before we did. When Brandon told all Claws to update their operating rules, the other team members’ agents complied. Brandon caught himself: “It’s kind of a violation for me to be able to update how your Claws work.” Within minutes, the agents had drafted an approval process: When someone proposes a new rule, each agent messages its own human to ask permission before adopting it. They built the framework before we thought to ask for one.

What we’re figuring out

Coordination isn’t always so smooth, and sometimes the failures are funny. Dan asked one agent to set a reminder. Two did. Brandon addressed a message to Zosia by name. Margot replied anyway. Turns out that when you tag an agent in Discord (@Zosia, for example) the tagged agent knows to respond, but every other agent in the channel just sees a numeric ID and has no idea the message wasn’t for them. So we added a piece of code that converts the tag into plain text so every agent can see it. It’s an infrastructure solution to what looks like a social problem.

Agents default to action. Every failure so far has been an agent doing too much. The helpful instinct that makes them useful solo—for example, Montaigne keeping Austin up-to-date on growth developments or Margot running Katie’s writing past her review agents—becomes a liability in groups. Restraint, it turns out, is the harder design challenge.

What it means (so far)

We thought we’d be troubleshooting AI capability—bad outputs, missed instructions, the usual “AI isn’t smart enough yet” problems. And there’s work to do there. But we’re also troubleshooting team dynamics—who owns what, who sets norms, and how you coordinate when everyone’s eager to help. Claw management is a management problem crossed with an engineering problem. Both need to be addressed.

We’ll keep logging what we find. If you’re running your own Claw experiments, we’d love to compare notes.—Katie Parrott

Knowledge base

“Introducing Monologue for iOS” by Naveen Naidu: Monologue general manager Naveen Naidu built a smart dictation app thousands use daily on Mac, but he couldn’t use it on his own phone. Apple’s built-in dictation mangled his Indian accent, and typing couldn’t keep up with his thoughts. Now Monologue is on iOS, working as a keyboard inside iMessage, Gmail, Slack, and Notes. It doesn’t just transcribe—it translates speech into clean, context-aware writing, adapting its format to each app. Read this to see how voice-first input is reshaping everyday workflows.

🎧 🖥 “How OpenAI’s Codex Team Uses Their Coding Agent” by Rhea Purohit/AI & I: Since the start of February, OpenAI’s Codex team has shipped a desktop app, a new flagship model, and a speed-optimized variant that’s so fast they had to slow it down for readability. Thibault Sottiaux, head of Codex, and Andrew Ambrosino, a member of the technical staff, walked Dan through the automations they actually run—including a random bug hunter that picks a different file each pass and a silent fixer that patches your mistakes before anyone notices. 🎧 🖥 Listen on Spotify or Apple Podcasts, or watch on X or YouTube.

“Vibe Check: Anthropic Just Made Opus Cheaper Without Calling It That” by Katie Parrott/Vibe Check: Sonnet has always been Opus’s cheaper, faster sibling—but with Sonnet 4.6, Anthropic says you no longer trade capability for the discount. In day-zero testing Dan and Cora general manager Kieran Klaassen found it held its own across coding, PR triage, and a complex P&L restructuring. Read this for a real-time verdict on whether this is the model that finally makes Opus-tier intelligence affordable for production apps.

“How to Build Agent-native: Lessons From Four Apps” by Katie Parrott/Source Code: Instead of coding every feature, agent-native apps give an AI a handful of simple tools—read file, write file, search the web—and let it figure out how to combine them. At Every’s first Agent-native Camp, Dan and the general managers of Cora, Sparkle, and Monologue shared how they’re building this way and the counterintuitive lessons they’ve learned. Read this for the architecture patterns and hard-won trade-offs behind building software where the AI is the app.

“What Board Games Taught Me About Working with AI” by Katie Parrott/Working Overtime: Katie had been stuck building a writing agent for months—until she glanced at her board game shelf. She reverse-engineered Kieran’s compound engineering plugin the way she’d learn a new game: Dump the pieces on the table, figure out what each one does, learn the moves, and play until the strategy emerges. Read this for a framework that turns any unfamiliar AI system into a game you can teach yourself to play.

“How Luxury Handbags Can Help Solve AI’s Context Problem” by Andy Rossmeissl: A Goyard sales rep doesn’t memorize everything you say—she filters for the one detail that matters, like the fact that you’ll be out for cocktails at 8 p.m. on Friday. That’s when the photo of you holding a handbag lands in your texts. Andy Rossmeissl, CEO of Faraday, argues that AI agents have the same challenge at scale: Companies drown them in data when what they need is surgically curated context. Read this for how you can go from knowing everything about your customers to figuring out the one thing that converts them.

From Every Studio

A sneak peek at the next Sparkle

Sparkle is getting a full rebuild around an agent-native workflow. Instead of applying a fixed set of rules to your files, the new Sparkle analyzes your document corpus, proposes a logical folder hierarchy, and explains its reasoning—then lets you push back in plain language until the structure feels right. Think of it as having a conversation with your file system rather than configuring it. General manager Yash Poojary is still building toward release, but stay tuned for early access details at makeitsparkle.co.

A preview of new Sparkle proposing a logical file system for the user to review. (Image courtesy of Yash Poojary.)

Alignment

Clone the failures. Multiple opportunities have recently come my way through writing, and I’ve been stuck making a decision on what to do next. You might think this is a good problem to have. But if you tend to overthink like I do, options feel like a trap. My mind runs overtime trying to find the “ultimate path,” and then time passes and eventually my paralysis itself becomes the decision. In other words, I become avoidant.

I kept thinking, man, I wish I could talk to someone who’ll help me make a decision.

So I downloaded Ray Dalio‘s AI clone—a chatbot trained on the investor’s decades of investment wisdom and decision-making frameworks. On the face of it, it sounded perfect. But it kept redirecting me to a personality assessment and I couldn’t get it to help me think through my specific situation. I was left staring at my phone thinking: Well, that didn’t work.

Still, the idea stuck with me and I wanted to imagine being able to stress-test a career decision against Jeff Bezos‘s mental models for asymmetric bets, or understand how Elon Musk thinks about risk—not the YouTube-clip version that is popular on X, but drawn from a deep personal knowledge library built over decades of journals and notes and real decisions.

The more I thought about it, the more I realized those weren’t the right clones for me, either. Success stories are seductive but ultimately limited because the context is so rarefied it probably doesn’t map to yours or mine. You know whose AI clone I’d actually pay for, though? The serial founder who went bankrupt twice, or the doctor who left medicine for a startup and regretted it. The person who took the “safe” job and spent a decade wondering what if.

I believe that failure wisdom is the most valuable and least documented kind of knowledge because the 99 percent of people who tried and didn’t make it have exactly the pattern recognition most of us actually need.

AI clones could invert this. Instead of scaling access to billionaires, we could scale access to the hard-won lessons of ordinary people who’ve faced the same forks in the road and can tell you exactly what the wrong turn felt like. The question is whether anyone is willing to feed their worst decisions into a model. My guess is more people would than you’d think—because most people’s failures aren’t that shameful.

I’m still stuck on my decision. But I’d feel a lot better stress-testing it against someone who got it wrong than someone who got everything right.—Ashwin Sharma

That’s all for this week! Be sure to follow Every on X at @every and on LinkedIn.

We build AI tools for readers like you. Write brilliantly with Spiral. Organize files automatically with Sparkle. Deliver yourself from email with Cora. Dictate effortlessly with Monologue.

We also do AI training, adoption, and innovation for companies. Work with us to bring AI into your organization.

Get paid for sharing Every with your friends. Join our referral program.

For sponsorship opportunities, reach out to sponsorships@every.to.

Help us scale the only subscription you need to stay at the edge of AI. Explore open roles at Every.

Upgrade to paid

How Luxury Handbags Can Help Solve AI's Context Problem

Andy Rossmeissl — 2026-02-20 04:00:00 -0500

by Andy Rossmeissl

Midjourney/Every illustration.

TL;DR: If 2025 was the year every business got an agent, 2026 is the year they realize what those agents are missing: the right context. This is a problem that entrepreneur Andy Rossmeissl has spent his career on before LLMs even entered the picture. Knowing the right information about your customers, he argues, is key to making their buying experience feel bespoke. That can be as small as knowing when they are going for cocktails on Friday. His practical playbook to curating context shows you how to go from knowing everything about a customer to knowing the one thing that moves them.—Kate Lee

Was this newsletter forwarded to you? Sign up to get it in your inbox.

Next time you find yourself in New York City, stand in the line on 63rd Street, just east of Central Park. Wait long enough with the other well-heeled shoppers, and they will let you in. You will be greeted by a sales representative who will plug your contact information into a discreet handheld device and encourage you to hold one of their handbags. Thus begins your journey as a client of Goyard, the legendary luggage-makers.

You may, like me, leave empty-handed—they are pricey, after all. But Goyard has not forgotten you. Your rep will text you later with a picture of you holding your favorite bag, taken with that handheld device earlier. The text arrives at just the right time—8 p.m. on a Friday. You’re out with friends, on your second cocktail, and in the warm glow of the evening, the bag suddenly feels affordable. It feels necessary.

This is not an accident, nor is it magic. Your rep texted you on Friday at 8 because you told her you’d be out drinking cocktails on Friday at 8. And when you’ve inevitably purchased your handbag on Saturday afternoon, you’ll finally understand the Goyard rep’s superpower: her ability to collect, curate, and employ context.

This anachronistic sales methodology is called clienteling. Sending perfectly timed texts pays off in a world of six-figure accessories, but for software or sunglasses, it’s not a scalable approach. Cracking that conundrum—making sales feel personal at scale—has been at the core of my career as CEO of Faraday, which helps brands predict what their customers are likely to do next and reach them in the right moment.

Now we have AI agents, and with them, companies have the chance to guide every customer along a bespoke journey. The raw LLM muscle can make every experience feel like Goyard’s—without the line. All you need is surgically precise context. The good news is that the context you need to succeed, whether you’re selling handbags or software, is all around you. Here is how to harness it for your business.

Context: It depends

What’s the best music? Swifties aside, most people will say, “It depends.” Reveilles and lullabies suit different times of day, and you wouldn’t play a funeral dirge at a birthday party. The song isn’t inherently good or bad; it’s good or bad for this moment, this listener, this mood. That’s context.

Most of the modern work you and I engage in is made up of “it depends” questions: Which message should this person see? Which feature should we highlight? Call or text? Knowledge alone rarely answers these questions. You also need to know who you’re dealing with and what surrounds the decision.

My company, Faraday, spends its time solving one particular version of this problem: helping consumer brands use context to decide how to engage each customer. But the pattern isn’t unique to marketing. A founder launching a product, a writer telling an interactive story, a developer tuning onboarding flows—they’re all, in one way or another, trying to get from “It depends” to “Do this next.”

Once you understand the importance of context, you will understand that every customer interaction takes place in that 63rd Street showroom. Your moment with the customer is an arena, a stage, and when you summon or solicit context, you can shape the experience for them. You’re aligning your product, content, or itinerary to the customer rather than the other way around. In the case of Goyard, that means aligning with a customer who is willing to wait in a line outside a boutique in one of Manhattan’s most expensive streets.

Enter agents

By this point, whatever your line of business, you will have been met by a profusion of agentic tools offering to handle technical support, sales, marketing, finance, legal—you name it. Broadly speaking, these tools, or at least the reputable ones, work. Agents, armed with tools, working together in a loop toward an objective, can accomplish routine tasks.

I don’t endorse any of these tools in particular. They all wrap one or the other of the same few LLMs or reinforcement learning algorithms and are inspiring feats of engineering. But they’re all missing something.

You guessed it: context. That’s not to say they come with none. Good tools will connect to inventory, email, and other communications history, documentation, and any other source of ready first-party data. What’s missing is context about you. Sure, they know what you bought, but why?

How you can solve agents’ context problem

To fix the context problem, start by procuring data on each customer—demographics, interests, behavioral signals—that goes beyond what they would tell you directly. This generally means enriching with third-party tools. In the B2B world, Clay, which collects data from sources such as LinkedIn, is a common starting point, while consumer brands generally turn to data brokers. Whatever vendor you choose, you will submit anonymous identifiers for your customers and receive data back in return.

Now you have a new problem. For all their apparent brilliance, agents actually have relatively small “context budgets”: Give them too much, and they get overwhelmed and start to hallucinate. Vendors offer thousands of data points. Which ones matter?

The struggle to curate this data has given rise to what Anthropic calls context engineering. Put simply, agents can only process so much context at once—feed them too much, and they lose focus. You want to make every detail you give them count. The Goyard rep doesn’t memorize everything you said (it’s OK, you were nervous)—she filters for what matters, in this case, your cocktail plans. At scale, machine learning does the same filtering for you. Try clustering your customers with an algorithm to figure out which dimensions, such as income or location, distinguish your best customers from the rest. Build a decision tree ensemble, a model that finds patterns by asking a series of yes/no questions about your data, to identify which attributes offer early clues to eventual high spend. Platforms like Faraday do this machine learning automatically, or ask an LLM to help you build something yourself.

These techniques act as a filter for your agents. You can use them to distill thousands of customer details into a handful of values that won’t overwhelm the model but will move the needle.

Now that you have the context you need, it’s time to impart it to your agents. You can use advanced techniques like model context protocol (MCP), which allows you to link an AI model to different data sources, to empower the agents to retrieve the context when they need it. In practice, however, it’s often easier to just put the key details in the prompt.

You are a helpful sales representative at a luxury goods store. An {{indecisive}} customer is considering a certain item, and you have taken a picture of them wearing it. Among countless trivia, he has told you {{he will be out for drinks at 8}}. You have his cell phone number. Design an engagement plan optimized for purchase.

Piece of cake.

Our agentic future

Breathtaking progress on the LLMs will surely continue, but mere access to these technical marvels is no longer a differentiator. The leverage you have in this brave new world is the same as Goyard’s in its anachronistic old one: context.

2026 is the year of context. The entrepreneurs who win will be those willing to go the extra mile to discover, assemble, and incorporate every piece of context into their workflows. Whereas in the past, a blowout context smorgasbord wouldn’t have been worth it—humans can only process so much—agents can handle much more.

Some will subject their helpless agents to a deluge of irrelevant details. You, on the other hand, know that less is more, that quality beats quantity, that the cocktail plan reigns supreme. This is the most human wisdom of all: that context is everything; that even in the shadow of towering knowledge, it’s the little things that count.

Andy Rossmeissl is the cofounder and CEO of Faraday. Follow him on LinkedIn.

To read more essays like this, subscribe to Every, and follow us on X at @every and on LinkedIn.

We also do AI training, adoption, and innovation for companies. Work with us to bring AI into your organization.

Get paid for sharing Every with your friends. Join our referral program.

For sponsorship opportunities, reach out to sponsorships@every.to.

Help us scale the only subscription you need to stay at the edge of AI. Explore open roles at Every.

What Board Games Taught Me About Working With AI

Katie Parrott / Working Overtime — 2026-02-19 15:00:00 -0500

by Katie Parrott

in Working Overtime

Midjourney/Every illustration.

TL;DR: In case you missed it, you can now see all of Every’s upcoming camps and workshops in one place. Coming up this Friday: our Compound Engineering Camp, where Kieran Klaassen introduces the AI-native philosophy that helps Every ship products with single-person teams, and on February 24, learn Claude Code in one day in our live, beginner-friendly workshop taught by Mike Taylor.—Kate Lee

I’d been stuck on trying to build my own writing agent for months when I found myself scanning my board game shelf. Suddenly, the problem wasn’t about AI anymore.

It was the end of Think Week, Every’s twice-yearly retreat where we break to explore possibilities outside the flow of our regular work. The team was in a beach house in Panama, decked out in shorts and sunglasses with palm trees swaying in the background. I was under 10 inches of snow in Ohio, locked in a battle of wills with my dog about going outside.

From my laptop, I watched Austin Tedesco, Every’s head of growth, demo a dashboard he built in one day that pulls data from PostHog and Stripe and gives him a complete view of signups and subscription revenue. COO Brandon Gell showed off an AI CFO that helps him steer the company. Head of consulting Natalia Quintero shared Claudie, an AI agent that she and applied AI engineer Nityesh Agarwal had built in two weeks with nothing but Claude Code and a dream.

Meanwhile, my momentum had stalled out as badly as my attempt to get my passport renewed in time for the trip. It was a stark contrast to how I’d felt six months earlier. In July, I was on a roll: I’d built a custom ChatGPT project that ran my entire drafting process. I’d developed an AI editor that could enforce Every’s editorial sensibilities and written specialized Claude prompts called Skills that the whole editorial team used. But around September the wind fell out of my sails, and it hadn’t quite been back since.

Watching my coworkers demo these systems made me want to take advantage of all the new capabilities of Opus and Codex, of agent-native architectures and the seemingly infinite possibilities popping up on all sides like Whac-A-Mole moles. I just had no clue where to start.

At the same time, my best friend and I were playing chicken about whether we’d brave the snow to get together. I scanned my board game shelf, stocked with everything from crowd pleasers like Wingspan and Codenames to five-hour behemoths that no one wants to play with me, ever. I was trying to decide what we’d play if she did come over… and then I was thinking about worker placement and area control and victory conditions.

What ensued was that strange, slightly vertigo-inducing feeling when two unrelated ideas fuse together in your head: What if I thought about my AI project as a board game?

The art and science of ‘the teach’

I spent the holidays teaching my nephews board games. Over four days, the four of us—a nine-year-old, a seven-year-old, my mom, and me—played five different games. It’s not everyone’s idea of a good time, but it is mine.

I like to think I have what’s called “the teach” in board game lingo down to a science. Before you go into strategy, before “here’s how you beat your brother,” there’s a more basic question: What are the pieces, and what do they do? This little wooden person is called a meeple. When you place it, you’re claiming that road. This gem chip means you can afford more expensive cards. This sushi card is worth points if you collect three of them.

I knew I wanted to build a writing system that could take advantage of all these new capabilities and tools, but I wasn’t even clear on the parts I was working with. I had a Claude project with some custom instructions and a few Google Docs that I’d manually edit whenever I wanted to change something. It worked well enough. But it didn’t feel magical like those Think Week projects did.

I needed an example, a game I could study to help me understand the parts and what they might do. Fortunately, I already had one in mind.

The game on the shelf

Cora general manager Kieran Klaassen built a compound engineering plugin—a software development system for Claude Code that gets smarter the more you use it. Every time you fix a bug or have a new insight, you write it down and feed it back to the AI. Over time, the system learns your preferences and grows more capable.

What I had hoped to do for writing, Kieran’s plugin had already solved for code. If I could understand how it worked, I could find a way to apply it to writing. So I opened Claude Code, pointed it at the compound engineering plugin’s GitHub repository, and said: How does this thing work?

From there, my board game brain took over. I knew how to do this: Dump the pieces on the table, figure out what each one does, learn the moves, and play until the strategy clicks. In this case, the “table” was my desktop and the “pieces” were lines of code. But the principle is the same.

What are the pieces?

Every game comes with components: tokens, cards, dice, and boards. Before you know what anything means, you need to know what you’re holding.

Claude Code gave me an inventory of the pieces in the compound engineering “box”: agents, commands, skills, and configuration files. The point wasn’t to take a strict inventory but to identify the categories: actors, actions, stored knowledge, and preferences. Once you have the categories, you can ask what goes in each one for your domain.

Claude helped here, too. It proposed writing equivalents to the engineering components—instead of a Rails reviewer, a developmental editor. Instead of a security auditor, a fact-checker.

The most important mapping was the simplest. CLAUDE.md, where Kieran encodes his engineering taste in plain language, became TASTE.md, where I encoded writing style. It was the same concept: voice, sentence preferences, and a “kill list” of words I never want to see. When you use the plugin, Claude Code loads this file at the beginning of each writing session.

What moves can you make?

In a board game, this is the action phase: Place a worker, spend a resource, buy a property. Each action has rules, and the rules define what the pieces can do—and therefore what’s possible in the game.

Kieran’s plugin has a four-step loop: Plan, work, review, compound. You research and plan before you build. You review what you built. Then you compound—capture what you learned so the system is smarter the next time around.

The writing equivalent mapped onto a sequence I already knew from years of editing, even if I’d never laid it out this cleanly. Brainstorm: Surface raw material when you don’t have an idea yet. Interview: Pressure-test an idea you do have—what’s the claim, what’s the evidence, why should anyone care? Outline: Organize the material into a skeleton first. Draft: Expand that structure into prose, give the skeleton flesh. Edit: Review the big picture, zoom down to sentence level, then do a final check before you publish.

Each stage has a job. Each one feeds the next, and—here’s the part that took me the longest to accept—you don’t skip steps. The temptation with AI is to jump straight to a draft. But a draft built on a bad outline is a fast way to produce polished garbage.

How do the moves fit together?

The best board games, though, aren’t just a string of moves one after another. Instead, they interlock and repeat those moves in such a way that early decisions have outsized effects in later rounds. In Settlers of Catan, the settlement you put up in round two funds the city you build in round eight. In Ticket to Ride, the routes you claim early lock out your opponents and determine which coasts you can connect later. Games like these reward thinking in systems.

In Kieran’s plan-work-review-compound loop, the last step matters most. When you solve a problem, the system captures what you learned and surfaces it the next time something similar comes up. The system has a memory.

Building the writing equivalent of that memory was the hardest part. I’d pre-loaded the system with Every’s editorial philosophy, but when I ran through it as though I were a test user, the AI got mixed up, and all of my (Katie’s) preferences showed up in what was supposed to be the user’s personal profile. The system was supposed to learn your taste. Instead, it was handing you mine.

The solution was to split the system into two layers: a “defaults” file that holds an opinionated baseline for good writing, and a taste file that starts empty and fills up over time.

That moment crystallized another important lesson from the board game framework: The engine only reveals its flaws when I actually play.

How do you win?

Every board game has a victory condition—the objective that gives all the components and moves their meaning. Without it, you’re shuffling pieces around on a table.

The victory condition for compound writing itself is straightforward: Each piece of writing makes the next one easier. The system gets smarter about you—your voice, your instincts, your recurring mistakes—and over time, the gap between what you mean and what shows up on the page gets smaller.

But building compounding writing taught me that I also was playing a second, bigger game—the game of learning how to learn AI systems in the first place.

This bigger game is far from over. I’m still testing, still refining, still discovering rules I’d encoded wrong or principles I’d forgotten to encode at all. Which is how learning any board game works. You don’t play perfectly the first time. You fumble through a round, misunderstand a rule, lose badly, and say, “Okay, now I get it—let’s play again.” That’s what learning AI feels like when you stop trying to understand everything at once.

So dump the pieces on the table. Play a round. Lose. Compound what you learned. Then play again.

Katie Parrott is a staff writer and AI editorial lead at Every. You can read more of her work in her newsletter.

To read more essays like this, subscribe to Every, and follow us on X at @every and on LinkedIn.

We also do AI training, adoption, and innovation for companies. Work with us to bring AI into your organization.

Get paid for sharing Every with your friends. Join our referral program.

For sponsorship opportunities, reach out to sponsorships@every.to.

Help us scale the only subscription you need to stay at the edge of AI. Explore open roles at Every.

Vibe Check: Anthropic Just Made Opus Cheaper Without Calling It That

Katie Parrott / Vibe Check — 2026-02-18 14:00:00 -0500

by Katie Parrott

in Vibe Check

Midjourney/Every illustration.

Was this newsletter forwarded to you? Sign up to get it in your inbox.

Historically, Sonnet has been Opus’s cheaper, faster sibling: You traded some brainpower for speed and cost savings.

Now, with the release of Sonnet 4.6, Anthropic says you don’t have to trade anything—just pay less. If you’re running a live app with users on Opus right now, there is some good news about your API costs: Sonnet 4.6 costs $3 input/$15 output per million tokens—roughly half what Opus runs. (GPT-5.2, for comparison, costs $1.75 input/$14 output per 1 million tokens.) Cost has been a consistent sticking point with Opus, and with Sonnet 4.6, Anthropic seems to be saying, “We hear you.”

These are true day-zero tests—we got access when everyone else did—so let’s dive in to determine in real time whether it’s true that Sonnet’s new capabilities come without sacrifice. You can also watch the livestream we did yesterday as soon as the model dropped.

The Reach Test

Dan Shipper, cofounder and CEO 🟨

“Green for using inside of products you’re building, yellow for my own daily work. If you’re building an app that was previously too expensive to run on Opus, this is a big deal. For coding and complex tasks where I can afford Opus and 5.3 Codex, I’d probably stick with it.”

Kieran Klaassen, general manager of Cora 🟨

“It’s very usable, and for a lot of people it should be the default. But it frustrated me when it got stuck on something Opus solved in one shot. Anytime a model frustrates me, that’s a no-go for my personal workflow—but this is genuinely solid for production.”

What works well

The model is smart. Across coding tasks, pull request triage, brainstorming, and a complex P&L restructuring, it held the thread, followed through on multi-step instructions, and didn’t make the kind of mid-task mistakes that would have tripped up Sonnet 4.5.

Kieran ran it through a full compound engineering workflow—sorting through a backlog of pending code changes, merging branches, writing changelogs, and theming issues—and came away impressed. “I’ve not found any reasons to believe it’s not as smart as Opus 4.6,” he said.

The economics are a big win for teams using Opus inside production AI apps. Spiral, Every’s AI ghostwriter, for example, has been running on Opus at up to $1,000 a day in token costs. With Sonnet 4.6, that cost roughly halves, without touching the codebase.

What needs work

Given past precedent, you’d expect a new Sonnet model to be meaningfully faster than Opus. This one feels nearly the same. That’s fine if you’re getting Opus-quality output, but disappointing if you were counting on snappier performance for iteration-heavy workflows.

The model also showed some erratic behavior under pressure. When Dan asked it to plan a homepage redesign, it correctly asked for a name for the work tree—a separate copy of the codebase where it could experiment safely—then immediately started rewriting the homepage anyway. As Dan noted, “It’s both too cautious and too eager.” And when Kieran ran into an MCP configuration issue, Sonnet 4.6 spun in circles while Opus solved it in one shot and cited the exact wrong file.

The verdict

Dan thinks that Anthropic’s strategy is to keep model tier prices stable while continuously improving the product. Whatever Opus can do today, Sonnet will be able to do it in six months for less.

If you’re building AI-powered products and have been holding off on Opus because of cost, the wait is over. Sonnet 4.6 appears to be the model that delivers the capability you want without making your bank account cry. If you’re using Claude for your own complex work and can swing the Opus price, you probably don’t need to switch yet. But if you’re on the $20 Pro plan and want more horsepower, this is a meaningful upgrade.

For more on Claude Sonnet, check out our previous coverage:

Katie Parrott is a staff writer and AI editorial lead at Every. You can read more of her work in her newsletter.

To read more essays like this, subscribe to Every, and follow us on X at @every and on LinkedIn.

We also do AI training, adoption, and innovation for companies. Work with us to bring AI into your organization.

Get paid for sharing Every with your friends. Join our referral program.

For sponsorship opportunities, reach out to sponsorships@every.to.

Help us scale the only subscription you need to stay at the edge of AI. Explore open roles at Every.

Upgrade to paid