A lot of people are saying "I would never hire this man". You're looking at it wrong. The idea is to find interesting problems that have unknown amounts of ROI

A lot of people are saying "I would never hire this man". You're looking at it wrong. The idea is to find interesting problems that have unknown amounts of ROI but should still be tackled by someone. Give him some cash and something interesting to chew and see what he comes up with. It's kind of like tending to a houesplant. If he produces nothing of value, then fine you know the idea was probably a dud anyway, but you didn't have to spend any time on it. And if he does spawn something of value, you can take the work and have more qualified engineers work on it full time. Personally, I find the best use case for people like this is to throw ethically questionable tasks at them. Stuff you shouldn't really have a full time employee doing, but would be perfectly fine outsourcing to a contractor who works off the books.