I actually like one-off requests from management. Not all requests, I try to deflect the easy/"easy" ones, explain why the hard/"easy" are hard, take the easy/"

I actually like one-off requests from management. Not all requests, I try to deflect the easy/"easy" ones, explain why the hard/"easy" are hard, take the easy/"hard" ones because it would be stupid not to and actively seek the hard/"hard" ones. To explain why "easy" problem is hard, the best way I found is to find a precedent. Like "remember that supposedly easy project that ended up costing so much money to the company, that's what you are asking me right now, I can do it, but don't expect a different result", or "No, I can't finish it by 10pm, last time it took 3 days, it will also take 3 days this time, if you want it to be done by 10pm, find someone better than me or reduce the scope". The goal, if you get the task, is to make it clear it is harder than it looks (bonus point if it isn't). With one-off tasks, you get to learn a lot in many fields. Being a jack of all trades is interesting too. "Interesting" work can become stale if you are doing it for too long. You may think that spending all day working on a state-of-the-art project that make inspiring YouTube videos is interesting, but it can quickly become routine. But if you are recognized as a wild card, you can get to work for some time on that fancy project, maybe just because there is a library that crashes or because they need you do whip out a quick UI, but you get to see the environment, work a bit, and see if it is as interesting as it looks, and if it is, and if you really saved their ass, you have your chance of getting in the team, especially if it really is your field of expertise.