AI code review
What is AI code review?
An automated first pass over a pull request, where a model flags bugs, security issues, style problems and missing tests, then comments inline before a human reviews. It does not replace the human reviewer. It clears the routine findings so people can spend their attention on design and intent. The most widely used tools now handle tens of millions of pull requests this way.
Why it matters
Two shifts make the automated pass more than a nicety. The first is volume. When agents and assistants generate code faster than people can read it, and some pull requests now arrive with no human author at all, a machine reviewer is often the only thing that looks at every line before merge. The second is consistency, since a tool applies the same checks at 9am and 5pm without the fatigue that lets a tired reviewer wave a change through. It does not replace judgment about design or intent, and it has a specific way of going wrong: flag too much low-value noise and people learn to scroll past it, taking the genuine findings with them. A review tool nobody reads is worse than none, because it looks like a safety net while catching nothing.
In practice
Every pull request gets an automated pass first, so by the time a person looks, the obvious null check and the leaked key are already flagged and the human can weigh whether the change is the right one at all. The tuning that matters is suppression. Cut the pedantic style nags hard, because the moment developers start reflexively dismissing comments, the one that mattered goes with them.
