Tell HN: The loneliness of a pretty good developer

Also, I'd try to seek out professional orgs and groups where you could be among people smarter than you - there's always people smarter than you, and even more knowing more than you in certain areas. Find the areas that could interest you - and seek out to meet people that know a lot, and learn from them. Bonus if it's really hard for you - you probably can use some stretching, in the intellectual sense. If you have some failures on the way - it's ok, as long as you learned something, even if it's your own boundaries and weaknesses.

On the other hand, try to delegate more. I know it would feel like you could do it better, try to ignore that and let others do it anyway. If they do it slightly wrong, try to just wait and see if they can figure out why it's wrong and learn from it. If they do it absolutely horribly wrong, try to suggest them better ways, but do not be overly pushy. Instead, try to figure out, as a personal task for yourself, why they are doing it wrong - do they just don't know a better way? Teach it to them then. Do they disagree with you on which way is better? Ask them to argue it and see if you can learn anything from their argument - even if it's only about whether they have different priorities and preferences. Try to encourage them to step up, even if initially it means sacrificing some efficiency.

It's reasonable to expect it to feel weird at first, especially if you're not used to or comfortable with working closely with others. The good news there is that that gets much easier with effort and practice, too.

It can be easy at first to want to just do stuff while people look on. You want to watch out for that; it can be gratifying, but doesn't much help anyone learn.

The key, I think, is instead to find the right balance between talking through your analysis of a problem and how you work through it, and giving whoever you're working with opportunities to practice doing the same with you as a sounding board and counselor.

It sounds basic, maybe, put like that. But it isn't; "problem" here can mean anything from "implement a function to add two numbers" to something that extends across an entire software stack and touches every level of infrastructure. It can be building an MVP from scratch. The point is that it's what your current mentee is ready for, or maybe just a little past that if you think they've got a habit of undervaluing themselves and could use a chance to build some confidence.

Above all else remember that your essential purpose as a mentor is to make yourself no longer necessary to the person you're mentoring. You are trying to engineer yourself out of this job, because to have done so means you've succeeded.

You needn't worry too hard about what comes next, I think. A good mentor never wants for work, and it's a surprisingly good way to keep your own skills sharp, because you can't teach like this unless you really understand what it is you're teaching.

Too, demonstrating you can work well with other people this way makes it very easy to continue into management, if that's what you want - and it is worth thinking about that, at our age. Some folks stay great engineers all their lives, but not all. And hell, after a few decades doing the same kind of work, some folks just find they want a change.

“If you are the smartest person in the room, you are in the wrong room.”

Are you making 7 figures? If not, you shouldn't feel guilty about how much you're working. It sounds like you could lean on your employer for a six figure raise which can pay for things like a personal trainer.