Senior Network Architect – HPC GPU Data Center

Advanced Micro Devices; San Jose (CA); Hybrid; USD 150,000 - 190,000; Full time

Job summary

Advanced Micro Devices (AMD) seeks a Senior Network Engineer to architect and operate high‑performance backend networks for large GPU clusters, including 10,000+ GPUs. You will own the fabric from GPU servers to leaf-spine, optimize RoCEv2 Ethernet fabrics, and model capacity for AI/HPC workloads.

You will collaborate with AI, data center, storage, and security teams, lead incident response, and mentor engineers while driving scalable, observable networking solutions.

Qualifications

• Significant experience designing, deploying, and operating production data center networks for AI/GPU/HPC environments.
• Experience building backend networks for GPU clusters containing ~10,000+ GPUs or similar hyperscale systems.
• Deep knowledge of data center networking fundamentals and modern leaf-spine architectures.
• Hands-on experience with RDMA, RoCEv2, PFC, DCQCN, QoS, and related technologies.

Responsibilities

• Architect, deploy, and operate high-performance back-end networks for large GPU clusters.
• Own end-to-end network path from GPU servers/NICs through leaf-spine fabric.
• Design RoCEv2 Ethernet fabrics and scalable topologies; model capacity and plan growth.
• Lead incident response, root-cause analysis, and optimization for network workloads.
• Collaborate across AI, data center, storage, and security teams to prevent bottlenecks.

Skills

RDMA

RoCEv2

Data center networking

Leaf-spine

GPU clusters

BGP/ECMP

EVPN/VXLAN

Prometheus

Grafana

Junos OS

NVIDIA/AMD ROCm RCCL

Education

Bachelor's or Master's in Computer Engineering

Tools

Juniper Junos OS

Prometheus

Grafana

RDMA tooling

Job description

Advanced Micro Devices (AMD) seeks a Senior Network Engineer to architect and operate high‑performance backend networks for large GPU clusters, including 10,000+ GPUs. You will own the fabric from GPU servers to leaf-spine, optimize RoCEv2 Ethernet fabrics, and model capacity for AI/HPC workloads.

You will collaborate with AI, data center, storage, and security teams, lead incident response, and mentor engineers while driving scalable, observable networking solutions.