More rules should mean better output. That's the intuition. I spent weeks building a comprehensive CLAUDE.md — 200 lines covering naming conventions, security rules, error handling, architectural patterns, import ordering, type safety requirements, and more. I was proud of it. I'd thought through every scenario. Then I scored the output. 79.0 / 100. My carefully crafted documentation was actively
I'm a software engineer in Japan. I've been using AI coding assistants — Claude Code, Cursor, Copilot — for about one years now. At some point I started keeping informal notes on how many prompt revisions it took to get production-quality output. After a few months, a pattern was hard to ignore. For tasks I described in Japanese: 4–6 revisions on average. 1–3. Same AI. Same model. Roughly similar
"Write a function to fetch the list of users." — same prompt, same codebase. Yesterday: getUsers(). Today: fetchUserList(). Tomorrow: loadAllUsers(). Six months of AI-assisted coding and I kept hitting this wall. My initial reaction was "maybe I need to write better prompts." I wrote better prompts. The functions got slightly better. New inconsistencies appeared elsewhere. The problem wasn't the A
At 3:17 AM on a Tuesday in Q3 2024, our production Kotlin 2.0 microservice fleet hit a 92% memory utilization threshold across 140 nodes, traced to a silent coroutine leak in Ktor 2.2’s request pipeline that had been bleeding 12MB of heap per second for 72 hours. We lost $14k in SLO credits before we found the root cause. A Couple Million Lines of Haskell: Production Engineering at Mercury (78 p