My advice, which I wish was plastered everywhere because I never found it: start and finish many small games first. Keep dreaming up and writing down the ambitious ideas, but leave those for another time.
Also, don't build your own game engine(unless you have plans to never ship). Learn to use Unity/UE4/GameMaker/etc.
Writing game engines is a blast, but they're the ultimate in feature creep/nerd sniping. You'll spend all your time building the engine rather than focusing on gameplay.
But I also think that for most small games (1-2 developers) it can also a be a good idea not to use an off the shelf engine either.
Write the absolutely minimum that you can get away with to achieve the gameplay and look you want to achieve. Hard code everything for your specific game.
It's true that if you want to write an engine you can distract yourself from focusing on gameplay. But I think you can actually fall into the same trap using an off the shelf engine too. You end up spending a pile of time learning and working with the advanced tools and not focusing on gameplay either!
The basic rendering APIs are actually not that hard to use if you forget about generalising everything for some future that never arrives, and there is a hell of a lot to be said for working on a code base that one person entirely understands. The development speed benefits of this should not be underestimated.
Yeah, but those 6 months you spent on getting the ocean waves in the background, for that 15 second cut scene, just right are the best 6 months ever. :)
