I can talk to someone for 20 minutes and know if they can be a good engineer on our projects. I don't need secondary signals. I'm willing to bet that anyone wh

I can talk to someone for 20 minutes and know if they can be a good engineer on our projects. I don't need secondary signals. I'm willing to bet that anyone who knows any area well can do the same thing. A construction worker can probably talk to someone for 20 minutes and tell if they can work on a construction site. It does mean that bureaucrats who don't actually know an area can't hire people for that area, but they've never been good at that. For qualified interviewers secondary signals are only useful at the initial filtering stage of deciding who to interview. For that I've found that some mix of (has gone to any college | has worked in the field | has side projects or hobby projects in the field) combined with just describing the job and asking them if they could do it is good enough. Then a quick 20 minute call determines if they're worth talking to more. Then a longer call with some questions and involving other team members, then maybe a test project and/or reference checks if possible.