When everyone in a game is cheating, you will lose by playing fairly. And like other commenters have suggested - everyone expects similar behavior.

> For context, I'm early in my career (3 YOE) and I don't deal with management that much yet (I'm still shielded by my tech lead and PM), so I'm always looking for advice on how to navigate these things. I really can't just say "YES, YES, YES" when I know very well that something won't be possible.

It depends on a lot of things, like how big the company is, how critical the project, culture, management style, etc. - but you'd be surprised at what is actually possible that may sound impossible. I've had those "this is impossible" thoughts a lot early in my career and more often than not I was wrong. The few times I wasn't wrong, I made sure to highlight risks I saw at the outset, and when the risks materialized, point them out at the retrospective or whatever for CYA purposes.

The things you do want to avoid at all costs are death marches, impossible deadlines with consequences at the end of it, maintenance slogs which will tie you to a very annoying treadmill for your entire tenure - it's hard without experience to know where these things lie, but if all else fails, you can do what everyone else does, try the impossible thing, and either surprise yourself, or when it becomes clear it won't happen, toss it to some overly ambitious sucker (I don't condone this, I just have seen it happen far too often).