Join NVIDIA, a pioneer in GPU innovation that has transformed PC gaming, computer graphics, and parallel computing. As AI research accelerates globally, the demand for scalable, massively parallel computation is soaring, an area where NVIDIA GPUs excel.

This role focuses on architecting and designing advanced multi-GPU systems for next-generation AI datacenter platforms. You will explore and define architectures that seamlessly integrate GPU compute, high-bandwidth memory, in-package interconnects, and GPU-to-GPU communication fabrics. The goal is to deliver unparalleled AI performance, scalability, and resilience.

Architect multi-GPU systems for both scale-up and scale-out configurations, prioritizing AI performance, scalability, and resilience for the agentic era.

Define, refine, and evaluate future architectures for high-speed interconnects like NVLink and Ethernet, ensuring co-design with GPU memory systems and networking hardware.

Architect RDMA-capable hardware and optimize transport layers for large-scale AI deployments leveraging GPUs.

Explore and develop novel high-density rack-scale AI systems, incorporating hundreds or thousands of GPUs interconnected via copper and optical links.

Utilize and adapt system models, conduct simulations, and perform bottleneck analyses to inform critical design decisions.

Collaborate with GPU ASIC, compiler, library, and software teams to foster efficient hardware-software co-design across compute, memory, and communication layers.

A Bachelor's or Master's degree in Electrical Engineering, Computer Engineering, or a related field is required.

We seek candidates with a minimum of 8 years of experience in system design and/or ASIC/SoC architecture, specifically for GPU, CPU, XPU, or networking products.

A deep understanding of communication interconnect protocols such as Ethernet, InfiniBand, NVLink, CXL, and PCIe is essential.

Demonstrated ability to architect multi-GPU/multi-CPU topologies, with a keen awareness of bandwidth scaling, NUMA, memory models, coherency, and resilience.

Strong analytical and system modeling skills are crucial for performance, power, and resilience assessments.

Excellent cross-functional collaboration and communication skills are a must.