Gergely Orosz · 7 min read · Gen AI Tools
What is happening with code reviews?
Read the original on newsletter.pragmaticengineer.com →
AI-generated code has dramatically increased PR volume, prompting CTOs and engineering leaders to rethink how code reviews should work. The article surveys pragmatic patterns such as humans reviewing AI-generated reviews, blast-radius triage, focusing on before/after states, and producing smaller PRs, while noting there is no one-size-fits-all approach and that some teams still do full manual reviews.
Key Takeaways
- AI code review tools are increasingly used, with multiple vendors and even multi-agent setups reviewing PRs and suggesting fixes.
- Most teams keep humans in the loop by reviewing the AI-generated review comments rather than every line of code.
- Triage changes by blast radius to decide review effort, with low-risk changes foregoing heavy review and high-risk ones receiving it.
- Reviews often focus on the before/after states, plan/tests, and database schema rather than implementation details, and aim to produce smaller PRs.
- There is wide variation in practice and some teams still perform full manual reviews; complete removal of human review is not widely adopted yet.
Excerpt
One question haunting the minds of CTOs and heads of engineering whom I’ve been talking with, is how to deal with large quantities of code review which have only been growing now that AI agents generate most code at many tech companies. Since the end of 2025, it has seemed that the era of devs writing code by hand is over at startups and in Big Tech. AI agents work faster and generate more pull requests (PRs) than devs ever did, and the size of those pull requests is also increasing. Today’s article summarizes some approaches to code review at…