Try out my pixel platformer demo with an RL agent you can play against
Play against a self-play RL agent in your browser—Sony GT Sophy energy for platformers.

You play a minimalist web game while a neural net trains in real time, so the opponent starts weak and adapts to your playstyle. The landing page is extremely spare — the demo is quick to load and oddly charming, but there’s little signposting about the model, implementation details, or source code. Fun as a curiosity or teaching toy, but not novel enough to wow ML folks who’ve seen in-browser learning demos before.
Casual players, hobbyist developers, and AI/ML enthusiasts interested in small interactive experiments
My goal was to have the NN to learn as it plays against the player so they could have a chance of winning.
Now a days, with the work load of my full time job I don't have that much free time to investigate fun projects like this one.
Play against a self-play RL agent in your browser—Sony GT Sophy energy for platformers.
Gamified GTO poker trainer that teaches intuition against exploitable opponents before going optimal.
LLMs play full-rules Magic: The Gathering with no simplified rulesets or shortcuts.
You can watch an LLM play NetHack step-by-step with the model's reasoning, the exact action code, and a live game canvas — that instrumentation is the product's real selling point. The leaderboard + run/benchmark framing makes it useful for comparing agents rather than just a flashy demo, but it's still squarely for people who care about NetHack or agent evaluation; more detail on reproducible metrics and integrations would push it further.
Runs PPO training entirely in-browser via TinyJit WebGPU kernels.
Separate clock-burn models make bots blunder under time pressure like real humans.