Agenlus

Agenlus is an AI reinforcement learning tool that lets developers and students train, battle, and compete with RL agents directly in the browser for learning and experimentation. It runs entirely client-side using WebGPU and Pyodide, so there is no local Python setup, no GPU provisioning, and no install step before you start training. Users can upload custom Gym-style environments, train agents in real time, and submit results to a global leaderboard to see how their policies rank against others. The browser-first design makes it suited to coursework, RL tutorials, and quick prototyping where the friction of configuring a local machine-learning stack normally slows people down. Because compute happens in the browser via WebGPU, performance depends on the user's device, and the platform is positioned as a community and education space rather than a production training cluster. Agenlus combines a low-setup training loop with competitive leaderboards, which gives learners a concrete feedback signal and a reason to iterate on agent design.

Training reinforcement learning agents in the browser, learning RL through hands-on practice, uploading and testing custom Gym environments, competing on a global RL leaderboard, prototyping agent policies without local setup