Simulation has long been one of the foundational pillars at NVIDIA, and with the release of Omniverse, a new collaborative platform for modeling, simulation and rendering, we are looking to advance physics-based simulations even further. To this end we wish to explore multiple techniques for accelerating physic-based simulations, namely deep learning, GPUs and efficient data structures. Thus, we are looking for highly motivated and experienced researchers and developers that can join a team at NVIDIA to work in the intersection of physics-based simulations, deep learning, and computer graphics.
What you’ll be doing:
- Join a small but highly skilled team of developers and researchers that seek to improve the simulation capabilities in NVIDIA’s Omniverse.
- Explore ways to accelerating physics-based simulations by means of deep learning and GPUs.
- Bridge the gap between real-time physics and higher-fidelity simulations traditionally found in respectively game engines and VFX productions.
- Develop cutting edge algorithms and acceleration techniques for efficient simulations.
What we need to see:
- M.Sc. or PhD in Computer Science, Physics or equivalent field or equivalent experience.
- 2+ years of work experience.
- Expert in at least two of the following topics: physics-based simulations, deep-learning and/or computer graphics.
- Proficient in C++ (at least 5 years of experience).
- Deep understanding of applied mathematics, e.g. numerical implementations of calculus and linear algebra.
- Experience with independent research as well as collaborative software development.
Ways to stand out from the crowd:
- Experience with simulation of fluids, muscles, cloth, hair, soft-bodies, and rigid-bodies.
- CUDA or other graphics APIs
- Sparse and adaptive data structures
- Knowledge of simulation techniques like FLIP, MPM, XPBD, FEM, etc.