Except that reviewing has become just asking your agent what that piece of code does in english. Reading code has a bit of cognitive load, that we will probably drop it in coming years
How does this work for you?
Do you ask about specific functions ("what does this function do")? Then it sounds like more work than actually reading it.
Or do you ask about the whole PRs ("what does this PR implement")? Then how do you catch things like security vulnerabilities, bad design choices, unnecessary dependencies, wrong architecture, etc? Those are things that benefit the most from human understanding (nitpicking style was always the easy but unimportant part of code review)
Not about a specific function anymore, or maybe once in a while, when something looks off.
> what does this PR implement
Most of the times, the coding agent already has the summary, and asks me specific questions where it needs me.
it very rarely also happens that i catch it taking shortcuts instead of doing the "right" thing. this happens while i am taking a look at the summary of the work it just wrote.
most of the correction happens later on when i test the final output irl.