Location: Chennai
Experience: 6-8 years
Qualification: Bachelor's or Master's degree in Robotics, Computer Science, Mechatronics, Mechanical Engineering, Electrical Engineering, or a related discipline.
Open Positions: 01
Role Summary
The Robotics Simulation Engineer will design, develop, test, and virtually validate high-fidelity robotic simulation pipelines within the NVIDIA Isaac ecosystem. The role combines physics-based simulation, reinforcement learning, foundation-model integration, ROS 2 middleware, GPU containerization, and sim-to-real engineering to support reliable deployment of robotic applications in an advanced manufacturing environment.
Key Responsibilities
- Architect high-fidelity, physics-based simulation environments using NVIDIA Isaac Sim, Isaac Labs, and Omniverse.
- Create and maintain accurate USD-based robot, sensor, workstation, and factory-scene assets, including RGB-D, LiDAR, IMU, contact, and force-torque sensor models.
- Develop scalable reinforcement-learning pipelines in Isaac Gym or Isaac Labs using PyTorch, including reward design, policy architecture, parallel training, and performance optimization.
- Integrate and fine-tune embodied-AI or foundation models such as NVIDIA GR00T for manipulation, locomotion, and related robotics use cases.
- Design ROS 2 packages, nodes, lifecycle components, custom messages, action servers, and DDS communication interfaces between simulation and robot-control stacks.
- Develop modular Python automation, custom Omniverse/USD extensions, synthetic-data generation workflows, and reusable engineering tools.
- Build production-grade Docker images using CUDA, NVIDIA Container Toolkit, and compatible Omniverse runtimes for headless GPU execution.
- Implement domain randomization across visual, dynamic, friction, latency, payload, and sensor-noise parameters to improve model robustness.
- Establish software-in-the-loop and hardware-in-the-loop validation frameworks and perform system identification using real robot telemetry.
- Support deployment across local GPU workstations and approved cloud environments, with automated simulation and regression testing in CI/CD pipelines.
Required Technical Skills
- Strong hands-on experience with NVIDIA Isaac Sim and Isaac Labs; working knowledge of Isaac Gym and NVIDIA Omniverse.
- Advanced Python development skills and practical C++ experience (C++17 or later) for robotics and performance-critical components.
- Solid understanding of reinforcement learning, PyTorch, reward engineering, policy training, inference optimization, and experiment evaluation.
- Proficiency in ROS 2, DDS, URDF, robot kinematics and dynamics, motion/control interfaces, and distributed robotics software design.
- Hands-on experience with Universal Scene Description (USD), OpenUSD APIs, URDF and/or MJCF asset workflows.
- Experience with Docker, NVIDIA Container Toolkit, CUDA-enabled workloads, Linux, Git, and CI/CD automation.
- Knowledge of sim-to-real methods, domain randomization, sensor modeling, system identification, and validation methodology.
Preferred Experience
- Experience with NVIDIA GR00T or other vision-language-action / embodied-AI foundation models.
- Exposure to TensorRT, Warp, PhysX, synthetic-data generation, and performance profiling of GPU-accelerated simulations.
- Experience deploying simulation workloads on multi-GPU workstations or cloud platforms such as AWS, Azure, or Google Cloud.
- Background in industrial robotics, autonomous systems, digital manufacturing, or virtual commissioning.
Education and Experience
- Bachelor's or Master's degree in Robotics, Computer Science, Mechatronics, Mechanical Engineering, Electrical Engineering, or a related discipline.
- Typically 6-8 years of relevant engineering experience, including demonstrable ownership of complex robotics simulation or autonomy projects.
- A portfolio, code samples, technical publications, or documented project outcomes demonstrating simulation and robot-learning expertise is desirable.
Key Deliverables
- Validated Isaac Sim/Omniverse environments and reusable USD scene graphs.
- Reproducible RL training pipelines, trained-policy checkpoints, evaluation results, and model configuration assets.
- Tested ROS 2 packages, launch files, middleware interfaces, and controller-integration components.
- Portable GPU-enabled container images and automated build, simulation, and regression-test workflows.
- Sim-to-real verification evidence, benchmarking results, system-identification outputs, and deployment-readiness recommendations.
- Complete technical documentation, setup guides, architecture notes, SOPs, and structured handover materials.
Behavioral Competencies
- Strong analytical problem-solving and systematic debugging skills.
- Ability to translate engineering objectives into maintainable simulation and software architectures.
- Clear written and verbal communication with cross-functional engineering stakeholders.
- Ownership mindset, attention to quality, and discipline in configuration management and documentation.
- Ability to work independently while collaborating effectively in an agile, milestone-driven environment.
Success Measures
- Simulation fidelity, stability, reproducibility, and performance meet agreed acceptance criteria.
- Robot-learning and foundation-model workflows execute reliably and demonstrate measurable task performance.
- ROS 2 middleware and controls interfaces pass functional and integration validation.
- Containerized environments are portable, documented, and suitable for repeatable local or cloud execution.
- Sim-to-real risk is reduced through documented evidence, controlled testing, and actionable engineering recommendations.
Send your CVs to [email protected]
