A

Andromeda Cluster

On-demand GPU resources marketplace for AI

Cluster Architect

DevOps & InfrastructureConfirmed live in the last 24 hours
Full-timeNo salary listed

About the Job

Requirements

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

Responsibilities

  • Own reference designs for standard clusters using current and next-generation NVIDIA platforms.
  • Create customer configurations, bills of materials, and rack layouts based on each customer's workload, scale, and site.
  • Analyze customer training and inference workloads to set design tradeoffs across latency, bandwidth, and GPU density.
  • Design within fixed site power budgets, maximizing GPU density while reserving headroom for storage, networking, and services.
  • Technically review vendor quotes and bills of materials.
  • Define burn-in, performance, and acceptance tests that determine when a cluster is ready for service.
  • Validate power, cooling, and floor requirements against each candidate data center.

Desired Qualifications

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

About the Company

Andromeda Cluster

On-demand GPU resources marketplace for AI

Andromeda AI Inc. runs a GPU compute marketplace that connects companies with a global network of providers for on-demand AI training and inference. It routes jobs across hundreds of providers—data centers, cloud services, and idle hardware from mining operations—handling sourcing, benchmarking, contracting, and a single consolidated billing flow. The platform differentiates itself by aggregating diverse compute sources and delivering end-to-end management, including the Andromeda Pricing Index to show fair, region- and contract-specific GPU prices, appealing especially to startups and labs that need flexible access. Its goal is to increase compute liquidity for AI workloads, making high-performance GPUs easier to access and more affordable for AI development and research.

Company size
11-50
Company stage
Not listed
Total funding
Not listed
Headquarters
San Francisco, CA
Founded
2023

Recently Posted Jobs

Explore more opportunities at Andromeda Cluster

See more jobs