Had this already 3 times that interviewers ask me out about their Engineering problems, and how I would solve them. Initially I got good or very good feedback but then afterwards was rejected. It feels weird, also if I consider that they might do this with every person interviewed. Crowdsourcing knowledge through interviews, yay ;)
Seems far more appropriate to get toy problems, if that is needed at all, or to get scoped problems but then get reimbursed if it takes several hours.
when i was younger i was interviewing at SomeStartup(tm) in Austin TX. I had a really good conversation with their VP of Engineering, a genuinely nice guy who cared about his people. The last step in the interview was an assignment to write a fairly specific caching system. I hacked on it for a couple hours and got it working per their requirements, demo'd it, handed back the laptop, and everyone liked my work. After that, radio silence. I wonder if I got fooled into solving a problem for them. Granted, that particular problem has been solved a hundred times over so i doubt it. Regardless, i've always wondered what the real story was there. Maybe they were just being polite.
Seems far more appropriate to get toy problems, if that is needed at all, or to get scoped problems but then get reimbursed if it takes several hours.