The thing with building something for myself is that it's hard. Sure, I do have a hobby project that will, hopefully, make the world better at some point. But that's a hobby project. It doesn't help with this crippling loneliness the slightest. And then starting your own company is a lot of responsibility. Much more than I'm willing to ever take. Not only do you have to have an idea good enough and that there's market for, you also have to devise a way to make money, preferably without ruining your product both in the long and short term. And delegation. I don't like delegation, but running your own startup demands a lot of it. I'm such a kind of person that it's almost always easier for me to do something myself when I know exactly how it's done rather than to explain someone else what I want.

The software industry is trending towards dumbing everything down and making increasingly user-hostile design decisions. This is the exact opposite of my values. Then there's the purely technical side of things. It's becoming scarily common for software developers to not understand the abstractions they're using and the compromises they're making while choosing those abstractions. This leads to poor-quality, underperforming products. It's become so ingrained into the culture that people legit look at you weird when you tell them that they don't need react and 50 JS libraries to build a website and that using the Android APIs in an app directly is a perfectly okay thing to do.

So, there are diminishing opportunities for creating a startup, and most existing companies' values go against mine. I feel lost at this point. I truly have no clue where to go next and what to look for. Nonprofits maybe?

I stand at the same point, and like others of my generation I'm considering switching job and moving to something completely different.

But if I'm being honest, I have to remind myself that the industry was not any better when I joined. Actually, I initially refused to work in that field for the same reasons that you list, except the epitome of dumb and user hostility at the time was Microsoft and Oracle. What made it possible for me to work in the industry was the raising tide of free software. Contributing to the success of it was interesting and fun and meaningful.

Now this tide have receded. Open source has merged with the industry, it incorporated its culture and helps greatly with the dumbing down and the over production of junk software.

I'm waiting for the next wave to come. I'm not sure where it will come from. My best guess at the moment as the most important, potentially disruptive technology, is functional OS distributions pioneered by NixOS and perfected by GuixSD, that has the potential of becoming the Debian of the future and revolution the way we assemble and distribute software and put an end to that unprincipled era of containers.

Rust is also an interesting piece of tech that has the potential to promote correctness and optimization in a field obsessed with simplicity.

So if one had to look for a job right now I would encourage either evangelization in one of those technologies maybe. That, or to work in a farm and enjoy computers on the side.