Fluidstack
Compute Engineer, Deployment
$164K–$206K
New York, NY +4 · Hybrid · Full-time · Posted Aug 11, 2026
- Location
- New York, NY
San Francisco, CA
Austin, TX
U.S. Remote
Seattle, WA - Experience
- Experience bringing server or GPU fleets of hundreds of nodes or more into production; no years specified.
About the job
Own compute turn-up from facility availability to production, qualifying server and accelerator racks at scale through provisioning, testing, and hardware failure triage.
What you'll do
- Configure and qualify racks, including firmware baselines, BMC/BIOS setup, burn-in, and node- and cluster-level validation.
- Automate and operate the provisioning stack to scale qualification; troubleshoot failures and coordinate RMAs and vendor escalations.
- Partner with deployment and operations teams during data hall turn-up and support incidents on newly live capacity.
What they ask for
- Experience bringing server or GPU fleets of hundreds of nodes or more into production.
- Deep Linux and out-of-band management experience with BMC, IPMI, and Redfish.
- Automating hardware workflows in Python or Go.
- Data hall experience with racking, cabling, component swaps, and failure triage; willing to travel for turn-up windows.
Nice to have
- Kubernetes-based bare-metal provisioning; accelerator platform bring-up (NVIDIA, AMD, or custom).
- Burn-in and stress harness design; DCIM and inventory tooling.
Pay
USD $164K – $206K base salary
Tools
Linux, BMC, IPMI, Redfish, Python, Go, Kubernetes, NVIDIA
AI focus
The role turns up compute infrastructure for AI workloads, including GPU and custom accelerator platforms.
Model-written summary · the original posting is authoritative