I often wondered about this while running a quant group at a hedge fund. Once your domain is somewhat specific, you tend to understand your own action constrai

I often wondered about this while running a quant group at a hedge fund. Once your domain is somewhat specific, you tend to understand your own action constraints. We have this many guys with these particular skills (algos, networks, etc) and we can only consider things that might make at least some amount. And it needs to be done within some horizon. Amazingly, it's much easier to have ideas if you have more constraints. Did you ever get asked by your teacher to write a story about anything? I always found that totally impossible. But a short story with a single protagonist, 4 pages long, occurring during some major life event, that's a lot easier to get started on. We built a page of potential things to investigate. They were always things built around some observation someone had while becoming experts. So for instance I'd stared at swap rates for a long time and wondered about whether there were inefficiencies we could look at. Also there was a look at different ways to use options, another thing we were strong in. Everything that worked was something adjacent to existing knowledge. Ended up spinning out a currency fund that did quite well.