Physical AI Engineer

TD SYNNEX • Clearwater, FL

7d • Hybrid

About The Position

We are building real-world Physical AI systems where models interact with physical machines. This role is for a robotics engineer with a strong reinforcement learning (RL) mindset, someone who wants to train, evaluate, and deploy intelligent behaviors that emerge through interaction, not just perception, by building the virtual environments, generating the data that trains our models, and developing the policies that eventually run on real hardware. Day to day, you will design simulation environments, produce large volumes of labeled synthetic data, train and evaluate learned policies, and work with engineers across robotics, controls, and perception to close the sim-to-real gap. You will work hands-on with NVIDIA Omniverse, Isaac Sim, physics-based simulation, and foundation models. This is a builder role: fast iteration, scalable training, and direct transfer from simulation to physical robots.

Requirements

• 5+ years in robotics, reinforcement learning, simulation, or applied machine learning. Degree in Robotics, Computer Science, or a related field, or equivalent hands-on experience.
• Hands-on experience training robotic agents in simulation, on physical systems, or both.
• Strong background in reinforcement learning, imitation learning, or learning-based control, including domain randomization and curriculum learning.
• Proven experience building simulation environments in NVIDIA Omniverse, Isaac Sim, or Isaac Lab, or comparable GPU-accelerated simulation platforms.
• Direct experience generating synthetic data for model training, including sensor simulation, annotation pipelines, and large-scale dataset generation.
• Working knowledge of OpenUSD as a robotics engineer, including asset conversion into a simulation pipeline from formats such as URDF or MJCF.
• Production experience with at least one RL library: RSL-RL, RL-Games, skrl, or Stable-Baselines3.
• Strong Python and the deep learning stack, such as PyTorch or JAX, with the ability to build and scale training pipelines beyond a single workstation.
• Experience applying or integrating foundation models into robotics or decision-making workflows.
• Builder mindset with a track record of moving learning systems from experiment to deployment.

Nice To Haves

• PhysX schemas and physics tuning.
• MuJoCo Playground, NVIDIA Warp, or Newton.
• Omniverse Replicator or comparable synthetic data generation frameworks.
• World foundation models used for data augmentation and photoreal domain transfer.
• Vision language action models or multimodal policies.
• ROS 2 or comparable robotics middleware, real-time systems, or physics engines.
• Model-free or model-based RL at scale, including distributed or cloud-scale training orchestration.
• Training perception models such as detection, segmentation, or pose estimation on synthetic data.
• C++ alongside Python for real-time robotics systems.
• Experience operationalizing learned policies on physical robots in production environments.

Responsibilities

• Build and maintain high-fidelity, physics-accurate simulation environments in NVIDIA Omniverse and Isaac Sim for training, testing, and validating robotic systems.
• Generate synthetic datasets at scale, including sensor and camera simulation, domain randomization, procedural scene variation, and automated annotation such as segmentation, depth, bounding boxes, and pose. You own dataset quality, versioning, and delivery.
• Design and run reinforcement learning and imitation learning pipelines using simulation-generated and synthetic data.
• Train and tune policies for control, planning, navigation, and manipulation, with emphasis on robustness and sim-to-real performance.
• Define task curricula, reward functions, and evaluation benchmarks so policy performance is measured before it reaches hardware.
• Model sensors, actuators, and contact behavior, and debug simulation instability, non-physical behavior, and determinism issues.
• Drive simulation-to-real transfer through domain randomization, system identification, and validation on physical systems.
• Build reusable tooling, APIs, and documentation so the broader team can stand up new environments and tasks without deep simulation expertise.
• Integrate foundation models to support reasoning, task decomposition, and human-in-the-loop learning.

Benefits

• Elective Benefits: Our programs are tailored to your country to best accommodate your lifestyle.
• Grow Your Career: Accelerate your path to success (and keep up with the future) with formal programs on leadership and professional development, and many more on-demand courses.
• Elevate Your Personal Well-Being: Boost your financial, physical, and mental well-being through seminars, events, and our global Life Empowerment Assistance Program.
• Diversity, Equity & Inclusion: It’s not just a phrase to us; valuing every voice is how we succeed. Join us in celebrating our global diversity through inclusive education, meaningful peer-to-peer conversations, and equitable growth and development opportunities.
• Make the Most of our Global Organization: Network with other new co-workers within your first 30 days through our onboarding program.
• Connect with Your Community: Participate in internal, peer-led inclusive communities and activities, including business resource groups, local volunteering events, and more environmental and social initiatives.

What This Job Offers

• Job Type: Full-time
• Career Level: Senior