Andromeda Deployment Engineer at Andromeda manages on-site GPU cluster builds from hardware delivery through acceptance testing. The role directs integrators and contractors and ensures quality and documentation for handover to operations.
Responsibilities
Andromeda's infrastructure team designs, deploys, and operates large-scale GPU clusters for AI labs and compute providers. The Deployment Engineer runs cluster builds on site, from hardware delivery to ready-for-service. You direct integrators and contractors day to day, own the punch list, and hand over a validated cluster to operations. You report to the Head of Infrastructure.
On-site execution: Receiving, rack and stack, cabling, power-on, firmware, burn-in, and acceptance testing.
Vendor management: Daily direction of integrators and contractors against the build plan.
Quality: Inspection against design, punch lists, and closure before acceptance.
Handover: Documentation, as-built records, and a clean handover to the operations team.
Runbooks: Deployment runbooks and checklists that shorten the next build.
Qualification
GPU deploymentsSystems skillsLiquid coolingOwnershipTravelStrong Candidates May HaveRack-scale systemsAutomationIntegrator backgroundWhy You'll Love It HereScripting for provisioning
Required
GPU deployments: You have deployed GPU servers or clusters in production data centers.
Systems skills: Linux, BMC and Redfish, firmware management, and hardware diagnostics.
Liquid cooling: Experience commissioning direct liquid-cooled racks.
Ownership: You run a site without close supervision and escalate early.
Travel: You can spend periods on site.
Strong Candidates May Have
Rack-scale systems: GB200 or GB300 NVL72 bring-up.
Automation: Scripting for provisioning, burn-in, and inventory.
Integrator background: Experience at an OEM or systems integrator delivering large builds.
Why You'll Love It Here
Lead builds end to end and see clusters go from crates to production
Shape the deployment playbook from the first builds
Preferred
Experience with rack-scale systems GB200 or GB300 NVL72 bring-up
Scripting for provisioning, burn-in, and inventory
Integrator background at OEM or systems integrator
