projects05 entries
projects
Systems I built to answer a question I could not look up — mostly agents, reinforcement learning, and the infrastructure underneath them.
Trove
Cafe operations — live inventory, autonomous reordering, forecast demand.
Next.js · XGBoost · Supabase
Cotter
Compliance testing for learned robot policies — pytest, for robots.
Python · MuJoCo · Wald's SPRT
Hivemind
A Prometheus alert becomes a root-caused pull request.
Go · Kubernetes · LLM Agents
Multi Agent RL Simulation
Two ride-hailing fleets learn pricing on a real Manhattan network.
PyTorch · IPPO · SUMO
Rocket the Robot
A walking robot built from scratch, taught to move in simulation.
Isaac Sim · Jetson · I2C
Smaller experiments and the things that never made it this far are all on GitHub.
github.com/yih0nk