Cluster Architect

Cluster Architect

Location: San Francisco, CA / Remote (US) · Full-Time

About Andromeda

Andromeda is a market and infrastructure platform to buy, sell, and operate compute.

We believe demand for compute will grow exponentially. So fast that a handful of vertically integrated providers won't be able to scale across operations, capital, supply chains, and politics to serve it. The result is a massive wave of fragmentation, with AI factories of every shape and size coming to market to fill this demand. Our job is to enable all of that fragmented compute to flow through one platform, delivering reliable capacity to model builders, research labs, and inference providers when they need it.

We believe every spare electron should be made productive for AI and we're building the platform that makes that possible.

We sit at the center of three forces:

• Companies that need reliable, high-performance compute fast

• A fragmented global supply of GPUs across hyperscalers, neoclouds, and independent data centers

• Capital, risk, and operational complexity that most teams are not equipped to manage

When we succeed, trillions of dollars of compute will flow through Andromeda. Builders get capacity when they need it. Providers get a reliable way to monetize, operate, and finance infrastructure at scale. Capital gets an easy way to deploy, hedge, and underwrite. In five years, Andromeda won't just participate in the AI infrastructure market. We will shape it.

The Role

Andromeda's infrastructure team designs, deploys, and operates large-scale GPU clusters for AI labs and compute providers. The Cluster Architect owns our reference designs. You will turn customer requirements into complete cluster designs across compute, fabric, storage, management, and the power and cooling interface, and act as technical authority on every build. You report to the Head of Infrastructure.

What You'll Own

• Reference designs: Standard cluster designs for current and next-generation NVIDIA platforms.

• Customer designs: Configurations, BOMs, and rack layouts built from each customer's workload, scale, and site.

• Workload analysis: Analysis of customer training and inference workloads to set design tradeoffs across latency, bandwidth, and GPU density.

• Power-constrained design: Designs that fit fixed site power budgets, maximizing GPU density while reserving headroom for storage, networking, and services.

• Quote and BOM review: Technical review of vendor quotes and BOMs.

• Acceptance criteria: Burn-in, performance, and acceptance tests that define when a cluster is ready for service.

• Site fit: Power, cooling, and floor requirements, validated against each candidate data center.

What We're Looking For

• 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.

Why You'll Love It Here

• 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

Andromeda Cluster is an equal opportunity employer. We celebrate diversity and are committed to creating an inclusive environment for all employees. We do not discriminate on the basis of race, religion, color, national origin, gender, sexual orientation, age, marital status, veteran status, or disability status.