Working remotely or in a hybrid environment, the full-time Physical AI Engineer will build and evaluate high-fidelity physics-based simulation environments using NVIDIA Omniverse, focusing on training robotic agents through reinforcement learning and synthetic data.
Key responsibilities
Build and maintain simulation environments for training robotic agents in NVIDIA Omniverse / Isaac Sim
Design and implement reinforcement learning and imitation learning pipelines using simulation-generated data
Integrate foundation models to enhance decision-making processes and ensure reliable performance on physical robots
Required qualifications
5-8+ years of experience in robotics, reinforcement learning, or applied machine learning
Strong hands-on experience with reinforcement learning and imitation learning techniques
Proven ability to build and train agents in simulation environments
Experience with the NVIDIA ecosystem, particularly Omniverse and Isaac Sim
Strong Python programming skills for developing scalable training pipelines