I study CS at Cornell and do reinforcement learning research with Prof. Sarah Dean. I work on simulating and training robots.
Research
- Balloon navigation with Prof. Sarah Dean's group: a conditional diffusion model that generates realistic, effectively unbounded wind for training balloon-navigation RL agents (balloon-env-dev), and controllers that plan under forecast error (balloon-sim).
- Legged locomotion with Cornell Physical Intelligence: teaching our hexapod to walk in Isaac Lab (hexapod-locomotion) and an online AMP learner (hexapod-cupi).
Projects
- Drone racing: PPO in Flightmare, an IROS 2021 replication (flightmare-vm).
- Marine search and rescue at the West Point Robotics Research Center: YOLO on unmanned boats to find people overboard (SNR-yolo-moos-integration, report).
Previously: two summers as a software engineer at Bloomberg.


