[Hiring] Member of the Technical Staff - Platform @Andromeda Cluster

Role Description

We are looking for engineers to build and operate the control plane that runs our fleet. Your responsibilities will include:

• The automated systems that take a cluster from bare machines to customer-ready
• Machine lifecycle between tenants: join, wipe, verify, rejoin
• Operating Kubernetes and Postgres across the fleet
• Contributing to our custom Kubernetes operators
• Scaling clusters from tens of nodes to thousands
• Participating in on-call rotations

Qualifications

• Impressive technical work you can go deep on, with impact in the world. That can take three years or twenty.
• 2+ years of on-call experience for critical production services
• Deep Kubernetes experience
• Strong Linux fundamentals: kernel, cgroups, containers, networking, storage
• Experience operating databases, monitoring, CI/CD, and cloud infrastructure at scale
• Familiarity with fleet management and capacity planning
• Prior experience writing operators, or with GPUs, HPC scheduling, or bare-metal hardware is nice to have, but not required.

Company Description

Andromeda was founded by Nat Friedman and Daniel Gross to give startups the scaled AI infrastructure once reserved for hyperscalers. The first cluster filled almost instantly. The years since went into the platform that makes compute liquid: it deploys into foreign datacenters and turns the hardware it finds into clusters that leading AI labs train on.

• Today Andromeda operates compute for 80+ customers across 30+ capacity providers
• Tens of thousands of GPUs under management
• Billions of GPU-hours supported, on everything from A100 to GB300