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 travel a lot. And every single time, I'm standing in an airport Googling "do I need an adapter for Thailand" while my boarding group is already lining up. The existing tools for this are bad. They're buried in blog posts from 2016 with popup ads, or they're a wall of text that doesn't actually answer the question. I already had most of the data. My destination guides on Vientapps cover dozens of
An SSG benchmark across five React frameworks, from one thousand You're building a marketplace. Or a documentation site. A wiki, Five minutes. Ten. Twenty. Maybe an hour. Maybe a stack trace. You don't know in advance — and the public benchmarks won't tell So I built a benchmark for the gap. Five frameworks in a pnpm workspace, each rendering one dynamic /posts/[id] from a shared deterministic d
Have you ever looked at code you wrote six months ago and thought: "Who wrote this monster?"? Relax, it happens to all of us. In software engineering, writing code that a machine understands is the easy part. The real challenge is writing code that other humans (including your future self) can understand, maintain, and scale. This is exactly where Software Design Principles come into play. In this
Part 1 of 5 in The New Engineering Contract — what it means to lead engineers when AI is doing more of the coding. SWE-CI tested 18 AI models across 71 consecutive commits. Most broke something on commit 47 they'd already broken on commit 1. That's not an intelligence problem. That's a learning system that isn't learning. A paper made me uncomfortable this month. Not because of what it found about