Lots of other good info has already been posted so the one thing I'll mention is that eventually people get sick of the "mad scientist" (I know as it has happen

Lots of other good info has already been posted so the one thing I'll mention is that eventually people get sick of the "mad scientist" (I know as it has happened to me). It's great to try new things and live on the bleeding edge but at some point the people you work with are going to get tired of switching to the new hot thing. You can only re-do your architecture so many times before people start tuning you out or start to say "no", then what happens? You'll go back to doing the same thing you were doing such as adding new features (using the already existing tech), fixing bugs, scaling out, etc. That or you'll find a new place to work which, depending on how long you lasted, means you might lose that equity you were given. Honestly, if you want to play mad scientist and build cool shit you should be a contractor and build solutions for companies that come to you with problems. You get to build it, hand it over and then move on to the next new thing.