Drimva

NVIDIA

Robotics Research Engineer - Robot Simulation and Evaluation

$184K–$356K

US, CA, Santa Clara · Full-time · Posted Oct 27, 2026

Experience
8+ years of relevant industry or post-degree research engineering experience.
Education
PhD in Computer Science, Robotics, or a related field, or equivalent experience.

About the job

Develop simulation and evaluation tools for robot foundation models, translating robotics research into scalable NVIDIA Isaac Lab Arena capabilities.

What you'll do

  • Build simulation frameworks for training and evaluating robot foundation models on NVIDIA Omniverse.
  • Design and evaluate robot-learning algorithms with researchers, and lead technical architecture and evaluation standards across Research and Product.
  • Develop, test, deploy, and maintain simulation and robotics software, including integrating LLM and agentic workflows.

What they ask for

  • PhD in Computer Science, Robotics, or a related field, or equivalent experience.
  • 8+ years of relevant industry or post-degree research engineering experience in robotics, simulation, or robot learning.
  • Track record in software design and translating research prototypes into production-ready software platforms.
  • Proficiency in Python and development experience with deep learning software stacks such as Pytorch or Jax.

Nice to have

  • High-impact publications at top robotics or machine-learning venues; expertise in reinforcement learning and/or imitation learning.
  • GPU-accelerated physics simulation expertise, NVIDIA robotics ecosystem tools, or experience developing and maintaining open-source projects.

Pay

184,000 USD - 287,500 USD base salary (Level 4); 224,000 USD - 356,500 USD base salary (Level 5)

Tools

Python, Pytorch, Jax, NVIDIA Omniverse, PhysX, Isaac Gym/Isaac Lab, Newton, CUDA

AI focus

Training and evaluating robot foundation models, developing robot-learning algorithms, and integrating LLM and agentic workflows into simulation and robotics.

Model-written summary · the original posting is authoritative

First seen Oct 28, 2026 · Last verified Nov 5, 2026 (UTC)