Senior Developer Technology Engineer - Simulation Performance

NVIDIA•Santa Clara, CA

9d•$224,000 - $431,250

About The Position

NVIDIA is seeking a hands-on engineer to help partners and customers optimize demanding physics and robotics simulation workloads on NVIDIA platforms. The role operates at the intersection of algorithms, mathematics, GPU architecture, and production software, finding performance limits in real applications and translating findings into durable code and actionable feedback for NVIDIA engineering teams. The position is intended for someone with a demonstrated record of taking stewardship of advanced systems and substantially improving them.

Requirements

12 or more years of relevant experience.
A degree in Computer Science, Applied Mathematics, Physics, Mechanical Engineering, Computational Science, or a related field, or equivalent experience.
A record of delivering large, reproducible performance improvements in production numerical, simulation, graphics, CAE, robotics, or scientific-computing systems.
Expertise in modern C++ and GPU parallel programming.
Strong foundations in mathematics, physics, numerical methods, geometry, or performance modeling.

Nice To Haves

Deep experience with physics engines, numerical solvers, geometry processing, differentiable simulation, or large-scale batched simulation.
Experience with CUDA, Warp, Newton, PhysX, Isaac Sim, Isaac Lab, MuJoCo, JAX, or comparable accelerated-computing systems.
A history of carrying important technical work beyond the breakthrough through integration, maintenance, documentation, and enabling the success of other developers.
Publicly inspectable stewardship of a consequential open-source engine, library, standard, benchmark, or software framework, with implementation and technical leadership visible in the artifact.

Responsibilities

Help partners and customers establish and execute performance strategies for physics, robotics simulation, and computational-engineering workloads on NVIDIA GPUs.
Build rigorous benchmarks and performance models, profile customer applications from system architecture through GPU kernels, and identify limiting factors.
Work with partner developers to improve algorithms, data structures, memory layouts, and parallel execution.
Collaborate with the responsible NVIDIA engineering team when evidence points to a platform bottleneck.
Apply knowledge of reinforcement learning, robotic kinematics, simulation setup, and simulation validation to partner challenges.
Write and upstream production C++, CUDA, Python, and domain-specific GPU code into simulation platforms, partner applications, and reusable reference implementations.
Work across developers, NVIDIA architecture and software teams, and open technical communities to make improvements durable beyond a single engagement.