I worked for software/hardware companies for 21 years. Some were stable income for me, but nothing amazing. A few were startups where I worked hard but they didn't go anywhere (fortunately I earned a stable, market-rate income before they failed). One turned out to be reasonably life-changing, and I'm very lucky for that. On and off over that period of time I worked on open source projects, some that didn't go anywhere, some that have been successful.
Right now I'm building something to try to sell. I'm not going to take any VC money or pursue high growth. I'm not chasing the latest whiz-bang AI whatever. I'm building a fairly boring product (that is still fun for me to build), using the tech stack that I want, to the standards of quality that I want. Right now I'm working hard on it, but the goal is that, once it's "done" and has customers (fingers crossed I manage to attract paying customers), I'll be spending no more than 10 hours a week on it (and hopefully less on average). If it "only" gets to $500k/yr in revenue or so, after some number of years running it, I'll consider that a fantastic success. If I can run the entire thing on one or two VMs, and it never grows past that infra-wise, that's a fantastic success.
You can listen to both the angel and the devil and still control your own destiny, outside of the scraps your employer will throw at you. Who knows, though, maybe this idea of mine won't work. But I'm happy I'm giving it a try.