Skip to content
  1. Home
  2. AI & Machines
AI & Machines

Automated code review is best at the comments nobody wants to write

Six months of data from three teams: the tools rarely catch architecture problems and almost always catch the tedious ones.

Automated code review is best at the comments nobody wants to write
Automated code review is best at the comments nobody wants to write — XENVORA Photograph: Programming Code, One Idea LLC / Stocksnap (CC0) · Licence

We asked three engineering teams to tag every automated review comment as useful, noise, or wrong. The distribution was more consistent than any of them expected.

Where the value concentrates

Null handling, forgotten error paths, inconsistent naming and missing test cases dominated the useful bucket. Design feedback — the thing senior reviewers actually provide — barely registered.

That is a reasonable division of labour. Reviewers stop spending attention on mechanical issues and arrive at the human conversation with the small stuff already resolved.

The failure mode to watch is confident wrongness on unfamiliar internal APIs, which teams learned to spot within a couple of weeks.

Tarek Nour
Software & Platforms
Tarek Nour

Tarek writes about developer tooling, build systems and the unglamorous work that keeps platforms alive.

Reader comments 1 comments

Sign in to comment.

T
Test Reader 30 Jul 2026

Bench numbers or it did not happen.