Senior Gpu System Architect

NVIDIA

Not specified; not specified; not specified
Hybrid
Multi-gpu system topology architecture
Nvlink ethernet infiniband pcie protocols
Rdma roce transport offload architectures
The role involves architecting and designing multi-GPU scale-up and scale-out systems for next-generation datacenter platforms focused on AI and HPC

Job Summary

  • The role involves architecting and designing multi-GPU scale-up and scale-out systems for next-generation datacenter platforms focused on AI and HPC.
  • Candidates must collaborate with ASIC, compiler, and software teams to enable efficient hardware-software co-design across compute, memory, and communication layers.
  • The position requires deep hands-on experience with system-level fabric networking architecture and practical expertise in GPU-to-GPU communication fabrics.

Matching Summary

The role involves architecting and designing multi-GPU scale-up and scale-out systems for next-generation datacenter platforms focused on AI and HPC.

Salary

Not specified; Not specified; Not specified

Skills & Requirements

Must-have

  • Multi-GPU system topology architecture
  • NVLink Ethernet InfiniBand PCIe protocols
  • RDMA RoCE transport offload architectures
  • Hardware-software co-design expertise
  • NUMA memory models and coherency awareness

Nice-to-have

  • NIC or DPU architecture experience
  • Chiplet interconnect architecture expertise
  • 2.5D 3D package co-design background
  • Interposer design experience
  • Python SystemC modeling skills

Key Requirements

  • BS/MS/PhD in Electrical or Computer Engineering
  • 8 years or more of relevant system design experience
  • Deep understanding of NVLink Ethernet InfiniBand CXL and PCIe protocols

Work Rights

Not specified

Tailored Resume

Cover Letter