Staff Software Engineer - Robotics (Simulation)

General Motors is building robotics simulation infrastructure to validate manipulation, perception, and navigation before hardware testing. The Staff Software Engineer will own the simulation environment and roadmap for virtual commissioning and digital twin capabilities, supporting robotics, AI, and test teams.

Responsibilities:

• Build and own our simulation environment for manipulation, perception and navigation validation
• Model robot platforms with physically accurate dynamics, kinematics, and actuation
• Implement sensor models for RGB-D cameras, LiDAR and other supported sensors with realistic noise characteristics
• Build scene and facility authoring tooling, including import paths from CAD and existing plant models
• Integrate simulation into CI so scenario coverage runs on every release candidate
• Support the AI team with simulation for data generation and policy evaluation
• Drive the virtual commissioning and digital twin roadmap, including workcell import and collaborative authoring
• Quantify and document where simulation diverges from hardware, so results are used with a known confidence

Requirements:

• 6+ years of professional software engineering experience, with 4+ years building or operating robotics simulation
• Experience validating real robot behavior against simulation and closing the gap
• Experience supporting other teams as an infrastructure owner
• M.S. degree or higher in Robotics, Computer Science, Mechanical Engineering, or a closely related technical field
• Equivalent industry experience with demonstrated simulation depth can be considered
• Hands-on experience with a commercial or opensource simulator, including Process Simulate, Isaac Sim or Gazebo or comparable
• Strong understanding of rigid body dynamics, contact modeling, and numerical integration, including where solvers break down
• Experience with sensor simulation and rendering, including LiDAR and depth camera models
• Experience with GPU-accelerated or parallel simulation for large-scale data generation
• Experience with robot description formats: URDF, SDF, USD, and their conversion pitfalls
• Proficiency in C++ and Python
• Ability to judge how much fidelity a given validation question requires
• Willing to state when a simulation result should not be trusted
• Comfortable trading simulation speed against accuracy
• Ability to work effectively with remote teams across multiple time zones
• Treats simulation as a product with internal users
• Experience with simulation to real transfer and domain randomization
• Experience with virtual commissioning or digital twin projects in manufacturing
• Familiarity with Visual Components, Process Simulate, or comparable process simulation tools
• Prior open-source contributions to Gazebo, Isaac Sim, or the ROS ecosystem