Type: url
Original: https://martinfowler.com/rachels-ramblings/code-review.html
Status: partial
Summary
A September 2026 post by Thoughtworks CTO Rachel Laycock, responding to a panel disagreement with Brian Houck of DX, arguing that AI-generated code volume exposes code review as having been overloaded with jobs it is poorly suited to. She proposes shifting judgment earlier through pairing, team design sessions, trunk-based development, automated tests, static analysis and fitness functions, keeping human review for exceptional changes, and focusing on engineers understanding systems rather than diffs.