Most code review happens at PR time, after code is already written, tested, and context is gone. With more AI-written code, tests pass but logic issues still s

Most code review happens at PR time, after code is already written, tested, and context is gone. With more AI-written code, tests pass but logic issues still slip through. Would earlier, pre-commit code reviews make more sense? I’m exploring this idea with a tool that reviews logic at commit time — curious how others think about timing reviews earlier.