Robotics AI Research Engineer - Imitation & RL for Humanoids

Agility Robotics; United States; Hybrid; USD 166,000 - 260,000; Full time

Benefits offered by this job

401(k) Plan

Equity: Company stock options

Medical, dental, vision insurance (100

Disability insurance

Unlimited PTO (Exempt)

Holiday and winter shutdown

Relocation assistance

Annual discretionary bonus

Job summary

Agility Robotics is seeking a seasoned researcher to help build and deploy next-generation robot foundation models and end-to-end policies for humanoids. You will design, train, and test imitation learning and reinforcement learning methods on real robots, running experiments with a cross-disciplinary team to translate research into production-ready systems.

The role focuses on locomotion, manipulation, and dynamic interaction in real-world settings, with collaboration across software and AI

Qualifications

• 3+ years of experience in learning-from-demonstration systems.
• Strong Python and PyTorch skills.
• Experience with DiffusionPolicy or similar tools.
• Experience collecting, training and testing robot data on hardware for manipulation tasks.
• MS in Robotics, CS, or related field; PhD is a bonus.

Responsibilities

• Design, train, and deploy policies for locomotion, manipulation, and dynamic interactions.
• Develop core RL infrastructure, including scalable training pipelines and evaluation frameworks.
• Design and implement new simulation environments for training and deployment.
• Research and implement imitation learning methods.
• Collaborate with Robotics Software and AI teams to productionize policies.

Skills

Python

PyTorch

DiffusionPolicy

Robotics data collection

Collaboration

Education

MS in Robotics or Computer Science

PhD (bonus)

Tools

Mujoco

Isaac Sim

Job description

Agility Robotics is seeking a seasoned researcher to help build and deploy next-generation robot foundation models and end-to-end policies for humanoids. You will design, train, and test imitation learning and reinforcement learning methods on real robots, running experiments with a cross-disciplinary team to translate research into production-ready systems.

The role focuses on locomotion, manipulation, and dynamic interaction in real-world settings, with collaboration across software and AI