> watching how someone solves a problem is just as important as the actual solution
I've seen this misnomer hundreds of times over the course of a 20 year career. It's a comfort-falacy. If someone interviews well they're going to give you a false sense of comfort and security in their ability. Interviewing well doesn't translate directly to working well. Interview processes in tech are overly hostile to the candidate - and I mean hostile in the sense that they apply unneeded pressure in condensed periods of time that don't reflect the majority of the work environment. Yes, we want people who can perform in crunch time, but that should not represent the bulk of work and if it does, there's another problem (I digress).
Another point that people seem all too eager to ignore is that how someone problem solves now is not as important as their potential or ability to improve and learn. A candidate that is excellent in interviews may not have the capacity for learning beyond what they're being run through in the interview. I've personally seen this happen at least a dozen times.
If one thinks a 30 minute coding exercise if giving them a good sense of a candidate's ability in both writing code and problem solving, they're part of the problem with hiring in tech.