Topics›AI Drift Prevention and Review Culture›Six Emerging Patterns for Code Review in the Age of AI-Generated PRs
★Very HighAct: HighFull content
Six Emerging Patterns for Code Review in the Age of AI-Generated PRs
The Pragmatic Engineer·September 8, 2026
New
Summary
Gergely Orosz surveys Anthropic, OpenAI, Duckbill Group, WeTravel, and Glide on how they're adapting review: (1) humans review the AI reviewer's comments, not the code; (2) risk-based triage by 'blast radius' (Duckbill nearly doubled PR throughput this way); (3) review the plan/tests/schema instead of the implementation; (4) enforce smaller PRs; (5) still review everything by hand; (6) no human review at all at some AI-native startups with heavy guardrails.
Why this signal
Exactly the systematic survey Neo's calibration flags as high signal — six named, attributable patterns from real companies that Thomas Engineering can directly compare its CodeRabbit-based approach against.
Original email
The Pragmatic Engineer · 9/8/2026
What is happening with code reviews?