AI Fixed My Code Before I Could

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

Knowledge base

"My AI Had Already Fixed the Code Before I Saw It" by Kieran Klaassen/Source Code: In the future, AI won’t just help you code—it will learn from every pull request, bug fix, and code review to create permanently improving systems. Kieran Klaassen calls it “compounding engineering,” and it's transforming how Every’s email assistant Cora gets built. Read this if you want to stop writing code and start building systems that build systems.

"Investing Secrets from Silicon Valley’s Best VCs" by Rhea Purohit/AI & I: Three of Silicon Valley's most influential VCs—Sarah Tavel, Mike Maples, and Nabeel Hyatt—have placed their bets not on raw technology, but on founders who can turn AI into magic. They're looking for the next wave of AI products that feel delightful, create new business models, and solve problems we don't even know we have yet. Read this if you want to understand how elite investors think about AI's future and why we should search for the “Japanese toilets of AI.”

"Best of the Pod: She Built an AI Product Manager Bringing in Six Figures—As a Side Hustle" by Dan Shipper/AI & I: Claire Vo created ChatPRD—an AI product management assistant now used by 10,000-plus product managers and generating six figures in revenue—all while keeping her demanding Chief Product Officer role. The kicker? She built it over a single weekend using AI tools. 🎧 🖥 Watch the full interview on X, YouTube, or listen on Spotify or Apple Podcasts.

"Sometimes the Best AI Solution Is the One You Don't Build Yourself" by Katie Parrott/Working Overtime: Katie Parrott was on a mission to create an AI-powered editor for Every. After five increasingly complex iterations—from individual Claude projects to full-on coding—she discovered the best solution wasn't a custom app at all, but Claude's new Teams feature. Read this for a refreshingly honest look at when to build versus when to use existing tools, and why the most elegant AI solution might be the one that requires the least engineering.

“Has AI Gotten Good Enough to Predict My Taste? I Had to Know” by Edmar Ferreira/Source Code We like to believe our preferences are as distinctive as fingerprints, but when entrepreneur in residence Edmar Ferreira fed six months of his browsing history to AI, the results were unsettlingly accurate. With just a paragraph describing his preferences, AI models could predict his content choices with up to 80 percent accuracy. Read this to explore the boundary between the predictable and ineffable aspects of human preference.

Work at Every!

We're looking for a lead social strategist to build Every's presence across the internet and architect how Every shows up online. You’ll own Every’s voice across X, LinkedIn, and wherever else the conversation is happening, plus manage accounts for our products (Cora, Spiral, Sparkle, Monologue). You’ll provide creative direction for our public narrative with the autonomy to shape it. This role is for someone who:

1. Is extremely online and knows how to make things spread—you’ve probably already made something go viral.
2. Loves to create content—written, video, and audio—and thrives across formats.
3. Ships fast with AI—you use Claude, Cursor, and whatever else to build MVPs, create content at scale, and automate workflows.
4. Is multidimensional—you’re into technology, code, writing, philosophy, design, and whatever else strikes your fancy.
5. Is multithreaded—you’re energized by juggling campaigns, experiments, and conversations across platforms.

You’ll have real ownership and creative freedom to do the following:

1. Architect the strategy, voice, and systems that define Every online.
2. Run bold experiments with room to break things, learn fast, and double down on what hits.
3. Extract the most compelling moments from long-form content and turn them into shareable formats that travel.

Your work will help shape the trajectory of the company: Think subscriptions, brand pull, product adoption, and conversation-shaping moments that move the business. You’ll know when something’s working (and why), and you’ll scale it with speed and rigor. You understand both how to market a subscription bundle and how to engineer distribution loops that compound. And you’re fluent in Figma, Descript, Buffer—or whatever gets the job done.

We want someone hungry who’s currently working in tech, ready to own this completely. You move fast, ship constantly, and—most importantly—get what makes Every special. If this is you, email Kate Lee at [email protected] with:

1. Your best work (tweets, posts, things you’ve made go viral)
2. Links to stuff you’ve built with AI
3. Links to things you like (your inspirations)
4. Why Every, why now

Collaborative filtering

Vibe coding for VCs. On September 19, join Bethany Crystal from Build First Academy, in partnership with Company Ventures, for an exclusive half-day workshop designed to teach VCs vibe coding—the art of building software with words instead of code. Participants at this in-person event in New York will learn Crystal's AI design framework, work through the AI no-code builder stack, and build their own applications with technical support on hand. Register through Build First or Luma.

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.

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.