Remote Senior GPU Cluster Architect for AI Infrastructure

Orion Placement; United States; On-site; USD 140,000 - 220,000; Full time

Benefits offered by this job

Bonus

Equity

Job summary

Orion Placement seeks a senior architect to own end-to-end GPU cluster deployments from customer requirements through deployment-ready design. You will craft configurations spanning compute, storage, networking, and supporting infrastructure, leveraging NVIDIA Reference Architecture principles.

Ideal candidates bring 7+ years in solutions architecture for GPU/HPC infra, hands-on Nvidia architecture experience, and expertise in InfiniBand, RoCE, and high-speed Ethernet.

Qualifications

• 7+ years of experience in solutions architecture, network engineering, or related roles for GPU/HPC/large-scale compute infrastructure.
• Deep knowledge of NVIDIA Reference Architecture and GPU cluster design principles.
• Hands-on experience with InfiniBand, RoCE, and/or high-speed Ethernet fabrics.

Responsibilities

• Own end-to-end technical architecture for GPU cluster deployments from customer requirements through deployment-ready design.
• Design GPU cluster configurations spanning compute, storage, networking, software, and supporting infrastructure.
• Translate client requirements into complete bills of design covering compute, storage, networking, and connectivity components.
• Apply NVIDIA Reference Architecture principles, including HGX and NVL72-based designs.
• Design high-performance network fabrics using InfiniBand, RoCE, and high-speed Ethernet based on workloads.
• Incorporate connectivity requirements (internet, VPN, firewall, circuit, protected optical) into architectures.
• Develop sparing strategies to meet contracted availability and SLA commitments.
• Adapt designs to site-specific power, cooling, space, and deployment constraints.
• Collaborate with data center, supply chain, deployment leadership, and program management to translate designs into build plans.
• Support acceptance test planning and define criteria to validate deployed architecture.
• Evaluate storage solutions like Weka, VAST Data, and DDN when appropriate.

Skills

GPU cluster design

NVIDIA Reference Architecture

InfiniBand networking

RoCE

High-speed Ethernet

Sparing strategies

Cross-functional communication

Tools

Weka

VAST Data

DDN

Job description

Orion Placement seeks a senior architect to own end-to-end GPU cluster deployments from customer requirements through deployment-ready design. You will craft configurations spanning compute, storage, networking, and supporting infrastructure, leveraging NVIDIA Reference Architecture principles.

Ideal candidates bring 7+ years in solutions architecture for GPU/HPC infra, hands-on Nvidia architecture experience, and expertise in InfiniBand, RoCE, and high-speed Ethernet.