Robotics Simulation Scientist — GPU RL & Sim-to-Real

Amazon; Bellevue (WA); On-site; USD 143,000 - 193,000; Full time

Benefits offered by this job

Health insurance

401(k) matching

Paid time off

Parental leave

Job summary

Amazon Robotics seeks an Applied Scientist to design, build, and validate GPU-accelerated simulation environments and policy training pipelines for manipulation and mobility tasks. You will develop RL and imitation learning workflows and transfer learned policies to real hardware, collaborating with SDEs, Technical Artists, and partner science teams.

You will own end-to-end simulation deliverables, contribute to design discussions, and help improve simulation fidelity and training methodologies

Qualifications

• PhD or Master's degree in a related field.
• 2+ years of experience with physics simulation platforms for robot learning.
• Experience training robot policies using reinforcement learning or imitation learning and evaluating them in simulation.
• Experience with articulated robot simulation, URDF/MJCF/USD formats and rigid/soft body dynamics.
• Familiarity with sim-to-real transfer concepts (domain randomization, system identification, physics calibration).

Responsibilities

• Design and implement GPU-accelerated RL and imitation learning environments in NVIDIA Isaac Lab for manipulation and mobility tasks.
• Build and maintain policy training pipelines supporting diverse models and evaluate trained policies in simulation.
• Characterize and reduce sim-to-real gaps through systematic validation against real robot data.
• Implement domain randomization strategies to improve policy robustness and transfer to real hardware.
• Develop sim-to-real transfer techniques including system identification and physics parameter calibration.
• Create robot embodiment validation tests to ensure digital twins match real hardware.
• Build data pipelines for recording, replaying, and augmenting demonstration data.
• Contribute to end-effector modeling and contact dynamics tuning.
• Author design documents for new simulation capabilities and participate in reviews.
• Collaborate with partner teams to understand model architectures and training needs.

Skills

Physics simulation experience

Robot policy training

Reinforcement learning

Imitation learning

URDF/MJCF/USD robotics formats

Sim-to-real transfer concepts

ROS2 robotics middleware

GPU-accelerated simulation

MoveIt/OMPL trajectory tools

Education

PhD or Master's degree

Tools

NVIDIA Isaac Lab/Sim

Omniverse

USD-based workflows

Job description

Amazon Robotics seeks an Applied Scientist to design, build, and validate GPU-accelerated simulation environments and policy training pipelines for manipulation and mobility tasks. You will develop RL and imitation learning workflows and transfer learned policies to real hardware, collaborating with SDEs, Technical Artists, and partner science teams.

You will own end-to-end simulation deliverables, contribute to design discussions, and help improve simulation fidelity and training methodologies