Boy Scout Rule in the Age of AI: Leaving Codebase Cleaner than You Found it
summary
Claims that 85% of developers use AI and that >40% of code is AI-assisted while only 28% trust that code, then prescribes concrete fixes: supply prompt-level context and memory files under 500 lines, enforce Prettier and linters and run them from agents, CI or PR hooks, generate tests as drafts then human-review (prefer TDD), treat E2E tests as brittle and harden them, restart AI sessions when refactoring, run CI tests and code review, and keep docs as versioned markdown plus Storybook/Playwright and OpenAPI/Swagger artifacts.
AI-written summary · about
chapters (11)
- 0:00AI and Boy Scout Rule
- 2:42Boy Scout Rule explained
- 4:30AI code problems
- 7:18Context for AI tools
- 10:27Memory files and prompts
- 12:18Formatting and linting
- 15:05Tests with AI
- 16:57Test-Driven Development with AI
- 18:04Refactoring with AI
- 20:09AI project docs
- 22:19AI: value or debt
related talks
37mAI-assisted development: Common pitfalls and smarter practices
46mCode Quality in the AI era, needed more than ever
26mBeyond the code: Leading engineering teams in the GenAI era
34mBuilding Guardrails for AI Coding Systems
01h 01m"Looks Good to Me": A Practical Guide to Handling AI-Generated Code
51mCode Smarter, Not Harder: AI-Powered Dev Hacks for All
49mClean Code in the AI era
39mAI-assisted development: Common pitfalls and smarter practices