I'd really advise against going down the 3D path first. Not just because the math gets harder, but there's all the additional overhead of assets. A 3D game wi

I'd really advise against going down the 3D path first. Not just because the math gets harder, but there's all the additional overhead of assets. A 3D game will require an environment, 3d models, etc. Maybe you can scrounge those, but unless he's really interested in the art aspect it's going to make it a lot harder. Just to get things going, try out a 2D game engine. Love2D is great if you're okay with Lua. Browser based engines have a great iteration loop. On that front Kaboom.js could be a really quick way to get something fun that you could iterate on or use as a protoptype (look past the silly assets, it's pretty well thought out). I saw some recommendations for Scratch, which might be fun for something simple, but it's really challenging (as far as I can tell) to scale up to non trivial games. Taking a step back, see what part of this motivates him. It's an awesome opportunity to channel interest into something creative, focus on that. It might mean cutting scope somewhere, but he'll get more out of it if you can mine that vein of interest.