AIPlaysMoonLander

A reinforcement learning PPO algorithm trained on a MacBook to play MoonLander.

This belongs to the fast-experiment track: media systems, game prototypes, developer tools, and playful interfaces built to test what newer AI and creative tooling can unlock.

The project is part of a broader pattern in Kyle's work: build the smallest useful version, learn from the behavior of the system, then use that prototype to decide whether the idea deserves more time, polish, or a totally different direction.