After 30+ years, I kind of reverse the pyramid, honestly. I look for readable, maintainable code, and try to make sure the writer is doing appropriate testing. But whether or not it works and is correct, that's not my job, that's the developer's job. I know developers that will spend days on a code review, which is a horrible waste of time. CI/CD and your deployment followup structure (canaries, monitors, gates, etc) should be catching any significant issues the vast majority of the time, and if it isn't, you need to spend time there.
My main concern as a reviewer is to make sure future people can understand the code when the inevitable modification comes along.
Things are different if the writer and reviewer are in a mentorship relationship, but even still, if you are only engaging with the code as a mentor when the PR hits, you're messing up the mentorship!
My main concern as a reviewer is to make sure future people can understand the code when the inevitable modification comes along.
Things are different if the writer and reviewer are in a mentorship relationship, but even still, if you are only engaging with the code as a mentor when the PR hits, you're messing up the mentorship!