AI code review
AI code review is using a model to examine a diff or a file for bugs, security problems, and quality issues before a human reviews it. It works best as a first pass that catches mechanical problems, not as a replacement for review by someone who knows the system.
What it is genuinely good at: unhandled errors, missing null checks, inconsistent patterns, obvious injection risks, tests that assert nothing. These are exactly the things human reviewers skim past on a long diff.
What it is not good at is judgement — whether the change is the right change, whether the abstraction will hold, whether it fits where the codebase is going. It also produces false positives confidently, which costs trust if you feed every one to the author.
What this means in practice
Reviewing AI-generated code with AI has a real blind spot: the same model that wrote a flawed approach will often endorse it. Use a review pass for mechanical checks, and keep human attention for whether the design is right.
Prompts for the tools this applies to
Related terms
- AI refactoring
AI refactoring is using an AI tool to restructure existing code without changing its behaviour — extracting functions, renaming concepts across a codebase, splitting large modules, or migrating between patterns.
- LLM evals
LLM evals are systematic tests of a model or AI feature's output quality: a set of inputs, a definition of a good response, and a scoring method.
- Test-driven development with AI
Test-driven development with AI means writing or generating the tests first, then letting an agent implement until they pass.