What are the responsibilities and job description for the Robotics Simulation Engineer position at Persona AI Inc?
We're looking for a Simulation Engineer to own and expand our NVIDIA Isaac Sim and Omniverse simulation infrastructure. You'll build the digital twin environments, sim assets, and CI/CD pipelines that our autonomy team relies on daily. You will help achieve the vision in which every robot behavior we develop is prototyped, tested, and validated in simulation before it runs on hardware.
As a simulation engineer at Persona, you will have an incredible opportunity to shape the simulation infrastructure that underpins our entire autonomy stack. You will report to the Behavior Coordination Lead and work closely with machine learning, locomotion, manipulation, perception, and devops teams.
Your Role:
Build, maintain, and extend simulation environments in NVIDIA Isaac Sim and Omniverse for humanoid robot development.
Develop and manage OpenUSD sim assets: robot models, tools, workpieces, customer facility environments (shipyards, steel fabrication plants, industrial sites).
Create and maintain physics-accurate digital twins of customer facilities (shipyards and industrial manufacturing) for behavior validation and deployment planning.
Integrate all autonomy subsystems (locomotion, manipulation, perception, hands, navigation, welding) to run against Isaac Sim.
Design and implement behavior-level CI/CD pipelines that automatically validate robot behaviors in simulation.
Prototype and mock-test robot behaviors (welding, navigation, manipulation) in simulation before hardware integration.
Develop synthetic data generation pipelines for training perception, manipulation, and navigation models.
We're Looking For:
BS, MS, or PhD in Robotics, Computer Science, Mechanical Engineering, or a related field.
3+ years of professional experience in robotics simulation, digital twin development, or physics-based simulation environments.
Strong hands-on experience with NVIDIA Isaac Sim and/or NVIDIA Omniverse.
Working knowledge of OpenUSD (Universal Scene Description): authoring, composition, referencing, flattening, and debugging USD stage hierarchies.
Proficiency in Python; comfortable writing simulation tooling, test harnesses, and automation scripts.
Experience with at least one physics engine: PhysX, MuJoCo, Bullet, or Drake.
Solid understanding of robot kinematics, dynamics, and articulated body simulation.
Experience developing or maintaining CI/CD pipelines for simulation or robotics software.
Familiarity with containerization (Docker) and DevOps practices for simulation infrastructure.
Ability to create and manage 3D simulation assets (meshes, materials, collision geometry, articulation definitions).
Preferred or Bonus Qualifications:
Experience with NVIDIA Isaac Lab for reinforcement learning environment setup and training.
Familiarity with NVIDIA Cosmos for synthetic data generation and scene variability.
Experience with ROS/ROS 2 and Gazebo simulation.
Experience with Unreal Engine or other real-time rendering engines for robotics visualization.
Background in manipulation simulation: contact modeling, grasp planning, dexterous multi-fingered hand interaction, and force/torque estimation.
Experience with motion capture data pipelines and motion retargeting to robot models.
Exposure to behavior trees, task planning, or autonomy stack architecture.
Knowledge of Omniverse extensions development and Kit SDK.