Cluster Architect

The role

The Cluster Architect owns reference designs and turns customer requirements into complete cluster designs. This includes compute, fabric, storage, management, and the power and cooling interface.

What you will do

• Create standard cluster designs for current and next-generation NVIDIA platforms.
• Build configurations, BOMs, and rack layouts from customer workloads, scale, and site.
• Analyze customer training and inference workloads to set design tradeoffs.
• Design for fixed site power budgets to maximize GPU density.
• Perform technical review of vendor quotes and BOMs.
• Define burn-in, performance, and acceptance tests.
• Validate power, cooling, and floor requirements against candidate data centers.

Required qualifications

• Designed and delivered GPU clusters of 1,000+ GPUs.
• Detailed knowledge of NVIDIA HGX and NVL72 systems and SuperPOD reference architectures.
• Working knowledge of InfiniBand and Spectrum-X fabrics.
• Working knowledge of parallel storage (VAST, WEKA, DDN).
• Working knowledge of Slurm or Kubernetes cluster stacks.
• Understanding of rack power, liquid cooling, and how facility constraints shape a design.

Nice to have

• Experience with NCCL, HPL, and MLPerf benchmarking and cluster acceptance.
• Delivered GB200 or GB300 NVL72 deployments.
• NVIDIA-certified architect or equivalent.

Benefits

• Healthcare, dental, and vision coverage
• 401(k)
• Unlimited PTO

About the company

Andromeda is a market and infrastructure platform to buy, sell, and operate compute. The company aims to enable fragmented compute to flow through one platform to serve model builders, research labs, and inference providers.