Hey @johnternus I think it’s time for Apple to have quality wednesdays. t.co/C0KuEErGhu t.co/O1qVzcOHWf

Apple please I am also begging you https://t.co/45XEUPHbjD https://t.co/gc0pueKFRB

really not liking the trend of code review becoming agent ping pong…

Print ad for @linear in the latest edition of Arena mag with @embeeargh.

honestly this is my favourite way to work through github/linear issues and prs though i think there is still room for improvement, and probably some hidden workflows i haven't found yet

Last week, we partnered with @Linear for our Lenny and Friends Afterparty! Huge thanks to everyone who came out, hope you all left well connected!

Human code review is a bottleneck

Open sourcing my personal autonomous build agent - t.co/ty6ZHmPX6C Move a card to todo on linear, and it automatically plans, builds, ships, redeploys. All while taking notes and interacting with you on linear. This has been powering many of my personal projects recently

A frontier AI lab gave us unlimited token usage for two months. Our engineers already used AI throughout the day, often juggling anywhere from 3 to 20 agents at once. We used the unlimited budget to push beyond everyday coding and test where else agents could accelerate the company. Where possible, we gave them straightforward tasks with clear right and wrong answers. We also tested what happened when the work required more judgment. Here’s what we tried, and how much value each experiment delivered: 5/10 - More PR review agents We had multiple agents examine every PR from different angles. Useful, but our existing Codex review caught most of the same issues. 7/10 - Codebase sweeps We added a `/sweep` command to search the entire codebase for bugs and performance improvements. It worked very well, especially when we pointed it at specific packages and gave it some direction. 8/10 - API documentation audit Agents read our entire API documentation and codebase, identified mismatches between the docs and implementation, and cleaned up typos and spelling errors along the way. 2/10 - Linear triage Agents reviewed every untriaged ticket and assessed whether it was actionable. I was really hoping this would create a self-improving loop, but either the Linear tickets were too vague or the correct action was too ambiguous. 3/10 - Trace optimization Agents combed through our traces for slow queries, but struggled to close the loop consistently. 10/10 - Refactors Agents finished all of our outstanding refactors that were mechanical or borderline mechanical. My trust in AI for this kind of work is nearly 100%. It is definitely more reliable than a human. 10/10 - Computer use for QA After every significant refactor, agents open the app and test the affected flows like a user would. It has been reliable enough that I no longer have to manually QA every change. The biggest lesson: unlimited tokens aren’t the bottleneck. The docs match the code or they don’t. The tests pass or they don’t. The query gets faster or it doesn’t. The refactor preserves behavior or it doesn’t. The moment the task required deciding what *should* be done, like triaging product tickets, performance fell apart. That’s the biggest limitation of AI today: it can do an enormous amount of work, but I still can’t trust it to decide whether that work is right. Unlimited tokens don’t solve the judgment and verification problem.

It’s underappreciated how much effort goes into deciding the top-1 thing to work on Usually in the evening I go over ~20 small tasks across slack “save for later” and notion boards, then pick the top 3 things to focus on the next morning It takes significant mental effort, and there’s still a real risk of missing the bigger picture and prioritizing the wrong problem Does @linear or any AI-first task tracker actually reduce the cost of deciding what to work on next?

Whatever the decision you were going to make based on these numbers - good luck @thsottiaux