Figure
AI Training Infrastructure Engineer – Humanoid Whole Body Control
About this role
Figure AI seeks an AI Training Infrastructure Engineer to build and scale the training and deployment systems for reinforcement learning-based whole-body control policies on humanoid robots. You'll own the simulation, data pipelines, and orchestration infrastructure that enables rapid iteration and deployment of robot capabilities across the fleet.
What you'll do
- Own and scale training infrastructure for whole-body control policies including simulation, data pipelines, and orchestration
- Design fast, reliable, and configurable systems for controls engineers to train policies efficiently
- Optimize cluster utilization and minimize downtime to accelerate team iteration cycles
- Evaluate and integrate physics engines and simulation environments to balance realism with training speed
- Build robust tooling for policy validation and deployment from training to real hardware
- Optimize hyperparameters and infrastructure to maximize training efficiency and model performance
What they're looking for
- Python and PyTorch production experience
- Physics simulation tools (PhysX, MuJoCo, Warp, PyBullet)
- Reinforcement learning and policy distillation
- Robotics dynamics and control systems
- ML training infrastructure and scaling
- Distributed systems and job scheduling
- Contact modeling and photorealistic simulation
- Hardware deployment of ML and control policies
Benefits
- Competitive salary range: $200,000–$300,000 annually
- Full-time position with comprehensive total compensation package
- 5 days/week in-office collaboration in North San Jose, CA
- Work on cutting-edge humanoid robotics and AI technology
- Role with direct impact on robot capability scaling
Opens the application — the Jobs AI extension fills it for you. Set up autofill
Opens the official application on the employer’s site. No login required.
Figure
Figure develops advanced humanoid robots powered by AI technology. The company is hiring engineers across mechanical design, firmware development, manufacturing, quality assurance, and security to build and refine its autonomous robotic systems.
View all jobs at FigureLikely interview questions
- Walk us through a time you built or scaled training infrastructure for ML or robotics. What were the biggest bottlenecks, and how did you optimize for cluster utilization and iteration speed?
- Describe your hands-on experience with physics simulation engines like MuJoCo, PhysX, or similar. How have you balanced simulation realism against training speed in your projects?