Built on Moving Ground

On June 12, Anthropic disabled Fable 5—the most capable coding model Every had worked with—after a U.S. government ban forced the company to shut it off for everyone. There was no warning or migration window. The models we rely on, it turns out, aren’t ours to keep: A company—or a government—can switch one off overnight.

That instability is the backdrop to our week. Senior applied AI engineer Nityesh Agarwal spent weeks building around Claude’s limitations, only to watch Anthropic’s new dynamic workflows solve the problem overnight—the cost, he says, of working at the frontier. Laura Entis and Katie Parrott report on loops (where you give an agent a goal and it works in rounds, reviewing each result and re-prompting itself until the work is good enough) moving fast from engineering into non-technical work. Head of growth Austin Tedesco built an overnight NBA simulator and reworked Every’s subscription flow the same way; Katie’s half of the piece is a playbook for what to do when a frontier model disappears. Mike Taylor tried something stranger, building an AI persona of GitHub COO Kyle Daigle so the real conversation could focus on what no public record could tell him—which you can hear on this week’s AI & I. And Stella Garber, co-founder and CEO of Hoop, rebuilt an internal tool to be agent-native—then started shipping it to customers. Thursday’s edition rounds out the week with the AI topics the team has given itself permission to skip, a workflow for treating your Slack bot like a coworker, and Mistral’s surprise seat at the G-7 AI table.—Kate Lee

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

Knowledge base

“I Interviewed an AI Version of GitHub’s COO—Then Spoke to the Real One” by Mike Taylor/Also True for Humans: Before interviewing GitHub COO Kyle Daigle at Microsoft Build, Mike Taylor ran his questions past an AI version of Daigle built from the COO’s public record. The simulation’s misses marked where public information ran dry. He spent the real conversation on what the model couldn’t know: the three machines Daigle codes across on weekends, GitHub’s automatic model router, and an agent loop Daigle uses to grade how he communicates. Read this for the pre-interview simulation technique.

🎧 🖥 “How GitHub Deals With 17 Million Pull Requests a Month” by Mike Taylor/AI & I: With agents flooding GitHub—commits jumped from 1 billion last year to a projected 14 billion this year—guest host Mike Taylor asks COO Kyle Daigle how the platform helps developers handle the surge without telling communities which pull requests to trust. Watch or listen to this for how the home of the world’s code adapts when everyone ships with agents. 🎧 🖥 Listen on Spotify or Apple Podcasts, watch on YouTube, or follow the discussion on X.

“How Anthropic Makes Claude More Reliable” by Laura Entis/Context Window: Anthropic’s new dynamic workflows let Claude Code write its own plan and run multiple subagents through a long task. Senior applied AI engineer Nityesh Agarwal felt it directly: Weeks of custom workarounds he’d built for Claudie, Every’s AI project manager, were suddenly obsolete. Read this for the Mini-Vibe Check and to see where dynamic workflows pay off.

“Loops for Non-Coders” by Laura Entis and Katie Parrott/Context Window: The loop—an agent that hits a stopping point, writes itself a new prompt, and keeps going—has crossed over from engineering into everyday non-technical work. Head of growth Austin Tedesco used the same setup that built an overnight NBA simulator to rework Every’s subscription flow and pricing page. Plus: Katie Parrott’s playbook for coping when a model you rely on vanishes, as Fable did—save your sessions, build while you have the window, and find which work needed the frontier model at all.

“We Built Our Own Agent-native Tool. It Overhauled How We Build Software.” by Stella Garber/Every: Stella Garber and two cofounders of Hoop, none of them engineers, built an internal AI tool to wrangle their scattered customer-call notes in an agent-native way. They handed the model a few tools and let it reason inside Slack rather than scripting a fixed sequence of prompts—an architecture that they now ship to their own customers. Read this for the agent-native build walkthrough.

Log on

Get hands-on with how Every uses AI. These are the live camps, workshops, and meetups where team members teach the workflows behind our work.

Upcoming camp

• Codex Power User Camp: On June 26, Dan Shipper and the Every team host a two-hour live walkthrough of the Codex power-user guide, including setup, workflows, and Codex-native app development. Learn more and register.

From Every Studio

Cora becomes a full email app

A rebuilt Cora launches in alpha next week: The new version is a standalone email client—plus iPhone app—that replaces your inbox instead of just layering on top of Gmail. Want to try it in an early but working state? Email Kieran Klaassen at [email protected] to join the alpha.

Monologue now works with Apple Shortcuts

Monologue’s new Shortcuts integration lets you capture ideas without ever opening the app. Assign Monologue to your iPhone’s Action Button and one press starts dictation—or trigger it from Siri, a widget, or the Home Screen. You can also chain it into automations that route your voice wherever it needs to go, like dropping a note into Notion or kicking off a draft email.

Alignment

The hand across the page. Juneteenth—the day Americans mark the end of slavery—was a few days ago. I’m halfway through James Baldwin’s semi-autobiographical novel Go Tell It on the Mountain, reading several pages each night as an excuse to keep my phone out of my hands before bed.

It’s the best thing I’ve read all year. Baldwin drops us into the shoes of his young Black protagonist, John, as he’s about to turn 14 and his family’s Pentecostal church prepares to hold an all-night prayer service. The story takes place in late winter, in 1930s Harlem, and reads like a time capsule of the era’s bubbling social tensions.

Baldwin’s prose is constantly surprising; his turns of phrase ring so true that they disarm you, open you to feeling what his character feels: “John’s heart was hardened against the Lord,” he writes. “His father was God’s minister, the ambassador of the King of Heaven, and John could not bow before the throne of grace without first kneeling to his father.” You come to that sentence braced for a question of faith and leave it understanding a boy who cannot reach God because he cannot get past his own father.

The element of surprise is the first thing stripped out when AI is handed the most personal things we have to express. A few weeks ago, I listened to an AI-written wedding speech, and the disappointment I felt in the moment has stuck with me. The groom had handed off the one chance to say, in his own surprising and unrepeatable way, what his partner meant to him. What came out instead were words that could’ve been written for anyone.

I don’t expect everyone to be (or even approximate) a Baldwin or Hemingway, or to deliver JFK-like speeches. But you don’t have to write well. You only have to tell in your own specific way what happened, because what is surprising, almost always, is simply what is true.—Ashwin Sharma

We’re hiring a managing editor

Every is looking for a managing editor to run the daily publishing engine behind our editorial work: owning the calendar and the 11 a.m. ET send, editing and packaging pieces, coordinating freelancers, and sharpening the AI-assisted workflows behind the newsroom. If you have sharp editorial taste, experience running a high-frequency publishing operation, and a genuine interest in editing with AI—and you’re in New York or a nearby time zone—apply here. Not the right fit? Explore Every’s other open roles.

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 [email protected].