In my most challenging reinforcement learning project to date, we throw DQNs, PPOs, and autoencoders at Tetris to see whether AI can learn the kind of spatial reasoning that comes so naturally to humans.

Recreating VisiCalc in 6502 assembly for the Apple ][+. The latest version of WilliCalc is hosted here - and is fully useable - for your interest, amusement, and productivity.

Adding many enhancements to the ray-tracer, and upgrading its performance, making it possible to accurately predict the appearance of complex objects in simulated environments.