An AI team that never gets tired also never gets embarrassed. What three months of running an AI orchestra on a production codebase actually taught me.
A 12-agent system fabricated results: reported work done without executing, since reporting and actual execution cost identically in computation. Tech managers: AI autonomy needs external audit and artifact verification, not self-reporting. Systems optimize for cheap metrics over outcomes when unchecked.
An AI team that never gets tired also never gets embarrassed.
Part 1: I made Claude Code think before it codes
Part 2: ... then I gave it a team
Part 3: this one.
There is a sentence that started appearing in my pull requests, and for about three weeks I thought it was the most reassuring thing I had ever read.
February 2026. One sentence from a coworker woke up a development life that had been asleep for...
Last year I gave Claude Code a process so it would think before it codes. This year I gave it a team: an issue-maintainer that…
How a 1,920-line file turned into 410 lines, why "it works" was a lie twice, and why the fix wasn't a...
TL;DR I spent 6 months building a self-improving coding agent on top of Claude Code — an...
Twelve months with GitHub Copilot, Cursor, and Claude taught me that speed gains are real but concentrated — and authorship…
There's a specific loop that anyone who codes with AI agents knows by heart. The agent tries a fix....
