AI tools are changing the skill distribution on engineering teams in ways most managers aren't ready for. The gap between "can produce code" and "can solve problems" is about to get very visible.
Most 1:1s are just standups with fewer people. The ones that matter follow a different pattern entirely.
I build side projects because I enjoy it, and it happens to make me a better manager. But let's be clear, it's not a requirement.
3 patterns I'm seeing, 2 predictions for the next 6 months, and 1 experiment you can run this week. Management experience might be the best predictor of AI success.