andromeda - Cluster Architect
Requirements
• Delivered clusters: You have designed and delivered GPU clusters of 1,000+ GPUs. • Platform depth: Detailed knowledge of NVIDIA HGX and NVL72 systems and SuperPOD reference architectures. • Full stack: Working knowledge of InfiniBand and Spectrum-X fabrics, parallel storage (VAST, WEKA, DDN), and Slurm or Kubernetes cluster stacks. • Facilities: You understand rack power, liquid cooling, and how facility constraints shape a design. • Strong Candidates May Have • Validation: Experience with NCCL, HPL, and MLPerf benchmarking and cluster acceptance. • Rack-scale systems: Delivered GB200 or GB300 NVL72 deployments. • Certifications: NVIDIA-certified architect or equivalent.
Benefits
• Define the reference designs our clusters are built on • Work across many sites, vendors, and platforms rather than one fleet • High talent density, high trust team shaping the compute market • Comprehensive benefits for you and your dependents, including healthcare, dental, and vision coverage, 401(k), and unlimited PTO
Apply in one click
Upload My Resume
Drop here or click to browse · Tap to choose · PDF, DOCX, DOC, RTF, TXT