AI Training Infrastructure Engineer – Humanoid Whole Body Control

Figure is seeking an AI Training Infrastructure Engineer to own the training and deployment backbone for RL-based whole-body control systems on humanoid robots. The role involves scaling simulation, data pipelines, and tooling to accelerate policy iteration and deployment. Based in San Jose, CA with in-office collaboration.

• Own and scale the infrastructure used to train whole-body control policies (simulation, data pipelines, orchestration, visualizations)
• Design fast, reliable, and configurable systems for controls engineers
• Ensure high cluster utilization and minimal downtime
• Evaluate and integrate physics engines (e.g., NVIDIA PhysX, MuJoCo, Warp, PyBullet)
• Optimize hyperparameters and infrastructure for training speed and efficiency
• Build robust tooling for training → validation → deployment pipeline

• Strong software engineering fundamentals with production experience in Python and PyTorch
• Experience building or scaling training infrastructure for robotics, control systems, or large-scale ML workloads
• Familiarity with physics simulation tools (e.g., NVIDIA PhysX, MuJoCo, Warp, PyBullet)
• Working knowledge of dynamics, controls, and robotics systems
• Experience with reinforcement learning, imitation learning, or policy distillation