GPU Cluster Architect (AI)

Location: Remote - United States. Applicants must be authorized to work in the country where they apply.

Salary: $184K–$318K OTE, including base salary and performance bonus. RSUs may be available at certain salary grades.

Company

Nebius builds a full-stack AI cloud platform for data processing, model training, and production deployment, with infrastructure spanning compute, storage, networking, and applied AI.

What you will do

• Architect scalable GPU cluster topologies covering compute nodes, InfiniBand or Ethernet interconnects, storage, and control planes.
• Model LLM training and inference workloads to evaluate latency, bandwidth, GPU density, and other architectural trade-offs.
• Design and validate low-latency, high-throughput interconnects including InfiniBand HDR/NDR and RoCEv2 at pod and data-center scale.
• Integrate storage for training datasets, checkpointing, and related workloads.
• Analyze monitoring signals to identify design issues and improve reliability.
• Collaborate with site reliability, networking, storage, and data-center engineering teams to operationalize and scale the architecture.

Requirements

• 5+ years of experience designing clusters.
• Deep understanding of modern GPU architectures, including NVIDIA and AMD.
• Experience with HPC interconnects, including InfiniBand and RoCE.
• Strong background in systems architecture, networking, and hardware reliability.
• Experience scripting automation and telemetry pipelines with Python, Go, or similar technologies.
• Authorization to work in the United States is required.

Culture & Benefits

• Remote work with up to $85 per month reimbursement for mobile and internet expenses.
• Company-paid medical, dental, and vision insurance for employees and families.
• 401(k) plan with up to a 4% company match and immediate vesting.
• Paid parental leave, including 20 weeks for primary caregivers and 12 weeks for secondary caregivers.
• Company-paid short-term, long-term, and life insurance.
• Opportunities for professional growth, learning, ownership, and collaboration on AI infrastructure projects.