Large Language Models (LLMs) are getting stronger, while AI agents are getting more autonomous. In many ways this is good; the tools can solve increasingly complex problems, work on larger codebases, and produce more code. But how do we make sure that the code being generated is correct, aligns with our coding standards, and is maintainable? These are the harder problems in software engineering - problems that require experience and discipline. When we have the capability to produce code faster, we also have the capability to make a mess faster, especially if we have no or little control over the output. But how does one control the output of a coding agent in the first place? The answer is simple…

When a developer works with an AI assistant, much of the work shifts from writing code toward reviewing code generated within the boundaries of their prompts and constraints. Reviewing a colleague's code builds on shared history, trust, and context. By contrast, today's AI assistants—still in rapid development—are highly variable from day to day. Trusting them is difficult, which makes the review of AI-generated code particularly draining…