Claude Code team: review now runs continuously 'like spellcheck'
Simon Willison · rss · 2026-10-17
Simon Willison quotes Addy Osmani, Morgan Westlee Lunt, and Qing Wang on how the Claude Code team does code review:
- Core idea: "did a human read the code?" is an imprecise proxy for "what could go wrong with this change, and which check—automated or human—is the right one to catch it?"
- Review timing shifted: it used to be a gate at the end (write code, open a PR, wait for a colleague); now review runs continuously, more like spellcheck — most changes have been reviewed and fixed by Claude at least once before a human looks.
- This catches routine mistakes and saves team attention for higher-level questions and scrutiny.
More from coding & agent
- Founder builds 'Rue', an AI bot to triage thousands of cold inbound DMs — soleio · 2026-10-17
- freeCodeCamp releases free course on Docker, CI/CD and production deployment — kuanhoong · 2026-10-17
- 3 months running an AI voice agent in a real business: what worked and what didn't — Admir-Rusidovic · 2026-10-17
- Quarkus LangChain4j sample repos updated, with side-by-side Spring AI comparisons — blaizedsouza · 2026-10-17
- deltabox-lite open-sources agent-native sandboxing with diff, commit and abort semantics — tianyin_xu · 2026-10-17
- Sample Spring AI repos updated to latest Spring Boot and Spring AI versions — blaizedsouza · 2026-10-17