No matter what you choose, build stuff and be around smart people.
In general don't start a startup you're not willing to work on for ten years.
... you can think about that for everyone you hire: will I bet the future of this company on this single hire? And that's a tough bar.
Investors will sort of like write the check and then, despite a lot of promises, don't usually do that much; sometimes they do.
If it works out, you're going to be working on this for 10 years.
The hard part is that this is a very fine line. There's right on one side of it, and crazy on the other.