San Francisco, California, United States
Remote only
Location: North America Remote / San Francisco, CA · Full-Time
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.
You will build and run the layer every cluster at Andromeda runs on. We run GPU clusters across many providers, sites, and hardware generations that we do not build ourselves, and leading AI labs train on them. Your job is to make that hardware behave like one reliable platform.
A bad node can kill a multi-day training run, and a slow handoff between customers leaves paid-for GPUs idle. You will build the systems that find failing hardware and repair it automatically. The same systems move nodes between customers fast, keep each customer isolated, and give every workload the scheduling and storage it needs.
The platform runs on Kubernetes on bare metal, and we write it in Go. You join the platform team that runs it today, write the software, and run it in production. You carry a pager and work with our product engineers, who build on your layer, and our solutions team, who support customers on it.
Node health: Detecting failing GPUs, links, and hosts, and fixing or replacing them before a customer notices.
Node lifecycle: Adding nodes, then wiping, verifying, and handing them between customers without leaking data.
Workloads: The scheduling and virtualization layers customer jobs run on, from long training runs to short-lived VMs.
Storage and isolation: Storage that outlives the node it runs on, and network and storage isolation between customers.
Scale: A platform that stays reliable as the fleet grows.
Impressive technical work you can go deep on, with impact in the world. That can take three years or twenty.
Kubernetes internals: You have written controllers or operators, not only deployed to a cluster.
Linux and bare metal: You have debugged hardware and kernel failures in production.
Systems programming in Go: You have shipped production services in Go, or in Rust or C, and can ship production Go within your first month.
Production on-call experience: You have carried a pager for a system customers depend on, and made it quieter afterward.
Production ownership: You have shipped a change, run it in production, and fixed it when it broke.
Cloud-only experience: You have only used managed cloud services and never debugged the host underneath.
You need a finished platform: Much of what you build here is the first version.
You want to manage: This is an IC role.
HPC schedulers: Slurm or similar, especially running on Kubernetes.
Virtualization: KVM, QEMU, VFIO, or SR-IOV.
Distributed storage: Weka, VAST, Ceph, or similar.
High-performance networking: InfiniBand or RoCE, including isolation between customers.
Out-of-band management: BMC, Redfish, IPMI, PXE, and firmware.
You work on the layer every training run depends on, on clusters of the latest NVIDIA GPUs.
We are a unicorn, and our customers are leading AI labs and data centers in one of the fastest-growing markets on earth.
Small senior team, a seat at the table with leadership, and impact measured in days, not quarters.
Competitive pay and meaningful equity, plus 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.
Want to know if this job is worth applying to?
