Support us by letting the company know you found them on our website.
Member of Technical Staff - Bare Metal & Fleet Provisioning
Apply To This JobJob Description
Prime Intellect builds the open superintelligence stack used by frontier AI teams. This role builds the systems that turn bare-metal GPU servers into reliable, production-ready compute, owning the machine lifecycle from discovery and provisioning through validation, upgrades, repair, and secure reuse as the fleet grows.
Responsibilities
- Build automated discovery, network boot, OS imaging, and configuration workflows for GPU servers
- Automate BIOS/BMC/NIC/GPU driver and firmware configuration with staged rollouts
- Build hardware inventory/lifecycle services and acceptance/burn-in tests for GPUs, memory, and storage
- Integrate provisioning and health checks with SLURM and Kubernetes
- Build observability, quarantine, and repair workflows with datacenter teams
What we are looking for
- 3+ years operating Linux servers or building bare-metal infrastructure automation in production
- Hands-on PXE/iPXE, DHCP, image provisioning, and Redfish/IPMI
- Strong Python or Go plus Bash, and experience with partial-failure-tolerant automation
Additional Information
Based in San Francisco, CA. Compensation is $150,000-$300,000 cash plus equity incentives.
The Company: Prime Intellect
Decentralized compute and RL post-training infrastructure for AI, backed by Founders Fund and NVIDIA.
Read more ->
Prime Intellect is looking for
People working in
Engineering >
With these skills
PXE/iPXE & Bare-Metal Provisioning, IPMI/Redfish, Python/Go, Ansible/Terraform, Kubernetes/SLURM, GPU Fleet Diagnostics
Willing to work
Full-time
With an experience level
Lead / Principal
Location
On-site • San Francisco, CA
Capacity
40 hours/week
Salary budget
$150,000-$300,000
Equity option
Equity incentives (amount unspecified)
New jobs every day on Telegram
Join the Working In Crypto channel for a daily job digest and a weekly market report. No noise.
VibeMyWay
Designed for solo founders and small teams, it saves time by delivering ready-to-publish marketing materials quickly without subscriptions—buy once and run in your AI editor.
Learn More