Sr. Simulation & Synthetic Data Engineer - Remote

Intuitive•Sunnyvale, CA

2d•$160,300 - $271,400•Onsite

About The Position

The Sr. Simulation & Synthetic Data Engineer will join Intuitive’s Future Forward Research — Synthetic Data team, which builds simulation and data infrastructure for autonomous surgical capabilities. This hands-on engineering role combines 3D simulation, machine learning, and robotics to create virtual surgical environments and data pipelines for training perception and policy models. The engineer will design simulation environments, generate and manage labeled synthetic datasets, and work with machine-learning engineers to improve sim-to-real transfer for robotic surgery. The role also includes modeling soft-tissue physics and instrument behavior, validating simulations against physical phantoms and real systems, and developing reusable tools and documentation that help the broader team create new simulation tasks. The posting describes the role as fully onsite at Intuitive’s Sunnyvale campus, with a day shift. Compensation ranges by region: Region 1 is $188,600–$271,400, and Region 2 is $160,300–$230,700. The position may be filled at a different job level depending on business need and the selected candidate’s qualifications.

Requirements

Strong software-engineering skills, including writing maintainable, tested code in Python, C++, and/or C#.
Comfort with Linux, Git, Docker, and GPU workflows.
Hands-on experience building in a physics simulator, game engine, or rendering/VFX pipeline; robotics simulators, real-time engines, and offline graphics pipelines are all relevant.
Experience producing data used to train an ML model, or working closely with those who do, and understanding how data quality and distribution affect model behavior.
Working knowledge of 3D and physics fundamentals, including coordinate frames, rendering, rigid-body dynamics, and contact dynamics.
Bachelor’s degree in Computer Science, Computer Graphics, Robotics, Electrical or Mechanical Engineering, Physics, or a related technical field, or equivalent practical experience.
Several years of experience building games, simulation, graphics, robotics, or data systems is described as likely experience, though the posting emphasizes demonstrated work over a specific number of years.

Nice To Haves

Advanced degree (MS or PhD) in a related area.
Deformable-object simulation, such as FEM, position-based dynamics, or differentiable physics.
Experience with NVIDIA Isaac Sim/Lab.
Surgical robotics simulation experience, such as ORBIT-Surgical or dVRK-based environments.
Sim-to-real techniques, such as domain randomization, system identification, privileged learning, or residual policies.
Experience with vision-language or vision-language-action models and how simulation data supports them.
Procedural or generative asset creation, such as NeRFs, Gaussian Splatting, or diffusion models.
Distributed compute at scale, such as Ray, Kubernetes, Slurm, or multi-GPU/multi-node systems.
Medical imaging or surgical video pipeline experience.
Publications or open-source work in simulation, graphics, robotics, or AI.

Responsibilities

Design and build high-fidelity simulation environments for surgical tasks.
Generate scalable synthetic datasets, including photorealistic imagery, segmentation masks, depth, optical flow, and kinematic state; manage dataset quality, versioning, and delivery.
Implement domain randomization and procedural scene generation to improve sim-to-real transfer.
Model deformable soft-tissue physics, tool-tissue contact, and instrument kinematics to match Intuitive’s platforms.
Partner with ML engineers to define task curricula, reward functions, and evaluation benchmarks.
Develop and refine sim-to-real transfer strategies and validate simulated behaviors against benchtop phantoms and real systems.
Build reusable tools, APIs, and documentation that enable the broader team to create tasks without deep simulation expertise.
Help establish simulation environments, synthetic-data pipelines, quantitative sim-to-real correlation, parallelized data-generation infrastructure, and a reusable task-creation framework, as described in the first-year success criteria.