Robotics Intern

NVIDIA is seeking dedicated interns to join our robotics team. You will develop novel approaches to agentic robot policies, world action models, robot policy training and evaluation, and sim-to-real transfer, working alongside researchers and engineers who build NVIDIA's robotics and simulation platforms.

Responsibilities

• Work closely with experts in robotics and machine learning to scope a challenging project that is both scientifically novel and relevant to NVIDIA's robotics roadmap.
• Design, implement, and evaluate new robot learning methods using GPU-accelerated simulation and, where appropriate, real robot hardware.
• Run rigorous experiments: build strong baselines, design evaluation protocols, ablate your ideas, and communicate clearly.
• Write clean, well-documented research code that other team members can build on.
• Collaborate with researchers and internal product teams to transfer your research into NVIDIA products and platforms.
• Deliver results as prototypes, patents, product contributions, and publications at top-tier venues.

Requirements

• Currently enrolled in a Ph.D. program in Computer Science, Electrical Engineering, Robotics, or a related field.
• Research experience in one or more of the following: agentic policies, world action models, robot policy training and evaluation, or sim-to-real transfer, demonstrated through publications, preprints, a thesis, or open-source projects.
• Strong Python skills and hands-on development experience with a modern deep learning framework (e.g., PyTorch, JAX).
• Strong written and verbal communication skills, including presenting research results to both experts and non-experts.

Nice to Have

• First-author publications at top robotics and machine learning venues (e.g., CoRL, RSS, ICRA, IROS, NeurIPS, ICML, ICLR, CVPR).
• Deep expertise in robot learning: reinforcement learning, imitation learning, or training and fine-tuning large vision-language-action (VLA) or world models for robotics.
• Deep expertise in GPU-accelerated physics simulation (e.g., PhysX, Isaac Gym / Isaac Lab, Newton, MJX), including building simulation environments, assets, or domain randomization pipelines.
• Experience deploying learned policies on real robot hardware (manipulators, humanoids, or mobile robots) and closing the sim-to-real gap.
• Experience with large-scale, multi-GPU training and data pipelines with contributions to open-source robotics or simulation projects.