GPU System Architect

NVIDIA NVIDIA · Semiconductors · Bangalore, India

NVIDIA is seeking a GPU System Architect to design multi-GPU scale-up and scale-out datacenter systems for AI and HPC. The role involves defining system architectures that tightly couple GPU compute, memory, and interconnects for optimal AI performance, scalability, and resilience. Responsibilities include architecting system topologies, defining high-speed interconnects, collaborating on RDMA hardware, using system models for analysis, and enabling hardware-software co-design.

What you'd actually do

  1. Architect multi-GPU system topologies for scale-up and scale-out configurations, balancing AI throughput, scalability, and resilience.
  2. Define, modify and evaluate future architectures for high-speed interconnects such as NVLink and Ethernet co-designed with the GPU memory system.
  3. Collaborate with other teams to architect RDMA-capable hardware and define transport layer optimizations for GPU-based large scale AI workload deployments.
  4. Use and modify system models, perform simulations and bottleneck analyses to guide design trade-offs.
  5. Work with GPU ASIC, compiler, library and software stack teams to enable efficient hardware-software co-design across compute, memory, and communication layers.

Skills

Required

  • BS/MS/PhD in Electrical Engineering, Computer Engineering, or equivalent area.
  • 2+ years or more of relevant experience in system design and/or ASIC/SoC architecture for GPU, CPU or networking products.
  • Deep understanding of communication interconnect protocols such as NVLink, Ethernet, InfiniBand, CXL and PCIe.
  • Experience with RDMA/RoCE or InfiniBand transport offload architectures.
  • Proven ability to architect multi-GPU/multi-CPU topologies, with awareness of bandwidth scaling, NUMA, memory models, coherency and resilience.
  • Experience with hardware-software interaction, drivers and runtimes, and performance tuning for modern distributed computing systems.
  • Strong analytical and system modeling skills (Python, SystemC, or similar).
  • Excellent cross-functional collaboration skills with silicon, packaging, board, and software teams.

Nice to have

  • Background in system design for AI and HPC.
  • Experience with NICs or DPU architecture and other transport offload engines.
  • Expertise in chiplet interconnect architectures or multi-node fabrics and protocols for distributed computing.
  • Hands-on experience with interposer or 2.5D/3D package co-design.

What the JD emphasized

  • AI and HPC
  • multi-GPU
  • system architectures
  • GPU compute
  • high-bandwidth memory
  • in-package interconnects
  • GPU-to-GPU communication fabric
  • AI performance
  • scalability
  • resilience
  • system-level fabric/networking architecture
  • hardware-software co-design
  • NVLink
  • Ethernet
  • RDMA-capable hardware
  • large scale AI workload deployments
  • GPU ASIC
  • compiler
  • library
  • software stack
  • compute
  • memory
  • communication layers
  • interposer
  • package
  • PCB
  • switch co-design
  • rack-scale systems
  • hundreds of GPUs
  • system design
  • ASIC/SoC architecture
  • networking products
  • communication interconnect protocols
  • NVLink
  • Ethernet
  • InfiniBand
  • CXL
  • PCIe
  • RDMA/RoCE
  • InfiniBand transport offload architectures
  • multi-GPU/multi-CPU topologies
  • bandwidth scaling
  • NUMA
  • memory models
  • coherency
  • resilience
  • hardware-software interaction
  • drivers
  • runtimes
  • performance tuning
  • distributed computing systems
  • analytical and system modeling skills
  • Python
  • SystemC
  • cross-functional collaboration
  • silicon
  • packaging
  • board
  • software teams
  • system design for AI and HPC
  • NICs or DPU architecture
  • transport offload engines
  • chiplet interconnect architectures
  • multi-node fabrics and protocols
  • distributed computing
  • interposer
  • 2.5D/3D package co-design

Other signals

  • GPU System Architect
  • multi-GPU scale-up and scale-out systems
  • AI and HPC