Build, implement, and optimize CUDA kernels for equivariant neural network primitives — tensor products, segmented polynomials, and triangle-based operations — targeting peak performance across NVIDIA GPU generations.
Develop and lead incident response and disaster recovery plans for pre-production clusters. Work with infrastructure, networking, storage, OS, firmware, and application teams to harden systems. Train peers and users on practical security standard methodologies.
Build and implement next-generation NvSci software to enable seamless cross-platform functionality and efficient integration with user applications, hardware acceleration libraries, and frameworks on various SoCs.
Design, develop, test, and optimize software for our next-generation SoCs. In both pre-silicon and post-silicon phases of execution. Review architectural performance bottlenecks for various system wide work loads.
Build, implement, and optimize CUDA kernels for core MD simulation primitives. Be responsible for the end-to-end delivery of GPU-accelerated simulation math to external partners and the broader MD ecosystem. Integrate simulation primitives into major MD engines.
Develop innovative high-performance processor and system architectures, focusing on the memory system and energy efficiency.
Designing, implementing, and optimizing direct sparse solvers for existing and future GPU architectures. Working with library engineers, QA engineers, and interns on all library development aspects from design to implementation and test to release and support.
You and your teammates will own every part of the engineering mission of cuEST - from idea to shipped CUDA-X library.
Define, design, develop and verify features for next-gen RISCV controllers on our GPU; collaborating with hardware engineers and fellow software engineers.
Showing 601–610 of 1,261 jobs
Show more jobs
Germany, DE
Canada, CA
France, FR