Principal Robotics Simulation Architect

AMD AMD · Semiconductors · Austin, TX · Engineering

This role focuses on architecting and advancing simulation solutions for robotic systems, with a strong emphasis on integrating AI/ML for reinforcement learning, synthetic data generation, and autonomous behavior training. It involves building scalable cloud-based simulation infrastructure and developing sim-to-real/real-to-sim pipelines to minimize the reality gap. The role also includes designing SIL and HIL testing frameworks.

What you'd actually do

  1. Collaborate with cross-functional teams to build and refine robust pipelines for transferring learned behaviors, sensor models, and physics parameters between simulation and real-world robotic systems, minimizing the reality gap through domain randomization and adaptation techniques.
  2. Architect and deploy scalable simulation infrastructure on cloud platforms, ensuring seamless parallel simulation runs, distributed training environments, and on-demand access optimized for robotic use cases with low latency and high performance.
  3. Provide technical leadership in integrating AI/ML frameworks into simulation environments, enabling reinforcement learning, synthetic data generation, autonomous behavior training, and intelligent environment modeling.
  4. Design and implement SIL testing frameworks to validate robotic software stacks — including perception, planning, and control modules — within simulated environments.
  5. Develop and maintain HIL testing setups that interface real hardware components with simulated environments, ensuring real-time performance validation, system integrity, and hardware-software compatibility verification.

Skills

Required

  • Robotics simulation stacks
  • digital twin development
  • sim-to-real/real-to-sim pipelines
  • cloud-based simulation infrastructure
  • AI integration within autonomous environments
  • ROS and ROS2

Nice to have

  • NVIDIA Isaac Sim
  • Gazebo
  • MuJoCo
  • Webots
  • Unreal Engine
  • PyTorch
  • TensorFlow
  • Isaac Lab

What the JD emphasized

  • minimizing the reality gap
  • sim-to-real/real-to-sim pipelines
  • AI integration
  • reinforcement learning
  • synthetic data generation
  • autonomous behavior training

Other signals

  • AI integration within autonomous environments
  • enabling reinforcement learning
  • synthetic data generation
  • autonomous behavior training
  • intelligent environment modeling
  • sim-to-real/real-to-sim pipelines