That's not too different than what I did. I left Google fall 2018, at the time thinking I'd write a video game. That morphed into doing research on 2D rendering, because building the infrastructure for that was more interesting and exciting than the actual game. For a while (not the whole time) it was pure unstructured time, no outside responsibilities. I rejoined Google a year ago with a research software engineer title, mostly because it was a better place to pursue this research than just continuing on my own.
Partly I'm posting this to support 'lapsedacademic's assertion, you can do good research in industry. It takes a certain element of luck to land a spot, but it also helps to be very passionate about the research itself.
I'm now writing a paper on an exciting GPU algorithm, but it's pure love, funding doesn't depend on it, it just makes a nice point on my list of accomplishments during perf time.
I'm curious, what are your options for starting a company based on the research you're doing at Google? What kind of ownership do they get (I assume it's not nothing)?
I thought about starting a company, but that's not where my energy is, it's doing the science. Everything I do is permissively licensed open source (though if I had a chance to work on awesome rendering in a proprietary product I would not turn it down).
I'll also add, I looked into academia and talked to some people about steps in that direction, for example seriously considering taking on teaching a class at UC Berkeley. But I decided that the daily grind of academia was not for me. I think I get to spend a larger fraction of my time actually working on my problems at Google than I would at a university.
A Zach-like with music synthesis as the core mechanic. I still think it's a good idea, but I don't really know how to write fun games (I think you have to write one or two to get enough experience) and I do know how to write 2D renderers, and increasingly feel able to claim I know how to program GPUs as well. I think it was Ivan Sutherland who said a good plan for research is to do something you think is easy and other people think is hard.
Partly I'm posting this to support 'lapsedacademic's assertion, you can do good research in industry. It takes a certain element of luck to land a spot, but it also helps to be very passionate about the research itself.
I'm now writing a paper on an exciting GPU algorithm, but it's pure love, funding doesn't depend on it, it just makes a nice point on my list of accomplishments during perf time.