Description
NVIDIA is hiring a GPU System Architect to design multi-GPU scale-up and scale-out systems for next-generation datacenter platforms supporting AI and HPC. The role focuses on GPU compute, high-bandwidth memory, in-package interconnects, GPU-to-GPU communication fabrics, RDMA-capable hardware, hardware-software co-design, system modeling, simulations, and performance tuning. Candidates need an electrical, computer, or equivalent engineering degree, at least eight years of relevant system or ASIC/SoC architecture experience, and expertise in interconnect protocols, RDMA/RoCE or InfiniBand, multi-GPU topologies, and distributed computing systems. The posting is marked hybrid and lists base salary ranges of 184,000–287,500 USD for Level 4 and 224,000–356,500 USD for Level 5, with equity and benefits.
