[Terretta]

> When we do performance reviews of engineers we take the time to comb over at least some of the PRs the engineer has written.

Wait, whoever is reviewing the engineer isn't already familiar through exposure to – if not participation in – flow of work?

[colinhowe]

In general, yes. I believe there's a difference between writing a performance review based on your memory/notes of how someone was doing 6 months ago vs getting into the weeds of a sample of PRs/specs and reviewing those from a performance viewpoint.

[Terretta]

For sure. It's why review should be as continuous as anything one wants to improve. Annual review is a relic from the gold watch era.