What do you mean? They just.. Do it? Nothing about an engineer precludes having good sense at other things.

Nothing about an engineer includes having good sense at other things, either.

In fact, many of the best engineers are profoundly bad at figuring out what to work on.

Sure fine but if the question is 'how does an engineer know what's good to do' the answer is, "get good". Ask questions, learn what's important, and think about more than code.

That's naive. A competent well rounded human is far more of an asset than an engineer who can't handle anything else. I would never want to work with an engineer who had no good sense about anything else unless I had to, like if there was some skill only they had.

Great soapbox; not at all how the real world of startups operates, however.

If you have a nonfounder engineer, that person for sure shouldn’t just be doing whatever it is they feel like doing. They need to allow the founders to drive.

Nobody claimed or believes that an engineer "should be doing whatever it is they feel like doing". You've misread what you're arguing against and assumed it is saying something it very much isn't.

"If you find an idea that you think is both valuable to the company and interesting to you" means just that; you. Not your partner, not your boss, not anyone else. You.

The comment is wrong, both in theory and in practice.

Engineers at small startups are generally founder minded. They’re like founders, and are passionate about the product and the problem. That’s how they get to decide/involve in the decision about what to work on

As an engineer at a startup, not sure that matters much.

The whole point is the founders have the eye, not as much so the engineers.