Code Review
A second pass by someone who did not write the code, looking for what the author missed.
Code review is a second pair of eyes on a change before it lands. With AI-written code, its value goes up, because the author often cannot explain the implementation and therefore cannot review it.
The review changes shape when the code was machine-written. The question stops being "is this well written?" and becomes "does this do what it claims, and what did the model assume?" That is a different review skill, and it is a faster one: you are checking behaviour against intent, not style against taste.
A useful habit is asking the agent to explain its own change in plain language, then reviewing that explanation against the diff. When the explanation and the diff disagree, the disagreement is the finding.
Related terms
Vibe Coding
Writing software by describing what you want and accepting the result without reading it closely.
Testing
The cheapest way to know a change is safe: a check that runs without a human reading anything.
Agent
A model with tools, a system prompt, and a loop that lets it act and observe results.
Human in the Loop
A person who reviews, approves, or corrects the agent's work at key points.
Regression
A change that made something worse. The reason to keep tests at all.
Want this working inside your team?
It starts with one conversation about your workflow.