Data Center Operations & Engineering Lead

Data Center Operations & Engineering Lead

Los Angeles, CA (downtown), on-site at a customer data center

About FarmGPU

FarmGPU is a NeoCloud: secure, high-performance GPU infrastructure for AI workloads, delivered through local data centers. We also run large GPU clusters for customers as their on-site operations and production engineering team.

We're hiring a Data Center Operations & Engineering Lead to run FarmGPU's on-site team at a customer's multi-thousand-GPU cluster in downtown Los Angeles: liquid-cooled NVIDIA Blackwell racks across two floors of a high-rise data center. You lead three onsite technicians and own the seam between the building and the cluster: power, cooling and network as they affect the GPUs, and the GPU servers themselves. When the facility has a bad day, you see it first, know what it means for the racks, and get the right people moving.

What You'll Do

• Lead the onsite team: set shifts, priorities and on-call for three onsite technicians, coach them, and keep the work visible to the customer and to FarmGPU.

• Watch power, cooling and network for the cluster: CDUs, manifolds and coolant levels, rack PDUs and branch circuits, air handling and the hall environment; set thresholds and alerts with our production engineering team.

• Know the servers: triage GPU, NVMe, NIC, optic and link failures with the technicians and our remote SREs, and tell quickly whether a fault is the node or the building.

• Work with the facility operator: maintenance windows, facility notices, escalations and post-incident reviews; make sure building changes are planned around customer workloads. The operator runs the building's plant; you monitor it and hold the plan together.

• Lead the physical side of incidents: cooling or power events, leaks, contamination; decide when to stop work and when to bring load back, and write the post-incident report.

• Keep the hall clean and healthy: contamination control, filtration, inspections and cleaning schedules.

• Plan capacity: per-rack power and thermal headroom before new racks or floors are energized.

• Own change control and runbooks: methods of procedure for rack power-cycling, cooling-loop isolation and refill, firmware and component work; train the team on them.

• Run the site: access lists, safety, contractor coordination and a weekly site report for the customer.

• Schedule

• On-site Monday–Friday, daytime hours, at a site that runs 24/7.

• Escalation point for after-hours physical incidents, on site within 60 minutes when needed.

• What You Bring

• 7+ years in data center operations and engineering or other critical environments, including time leading technicians or a site.

• A mechanical or electrical engineering background (degree or equivalent experience).

• Deep working knowledge of data center power (UPS, PDUs and RPPs, branch circuits, N+1 and 2N designs) and cooling (CRAH/CRAC, chilled water, CDUs, direct-to-chip liquid cooling).

• Hands-on GPU or HPC server experience: BMC/Redfish, DCGM and nvidia-smi, common hardware faults.

• Data center networking at the physical layer: fiber, optics, structured cabling and link troubleshooting.

• BMS/EPMS experience and comfort with dashboards and alerts.

• Incident command and clear written root-cause reports in a 24/7 environment.

• Safety leadership: LOTO, EHS and contractor oversight.

• Calm, direct communication with customers and facility operators.

• Ability to lift up to 50 lb, work on ladders and in a loud, cold data hall, with or without reasonable accommodation.

• Live within 60 minutes of downtown Los Angeles.

• Preferred Qualifications

• NVIDIA HGX B200/B300 or other liquid-cooled GPU racks at 40 kW and above.

• Contamination control in data halls.

• Experience in colocation or multi-tenant buildings.

• Certifications such as CDCP/CDCS, OSHA 10/30, or a PE or EIT.

• Why FarmGPU?

• Run one of the larger GPU clusters on the West Coast, with a team that operates GPU fleets every day behind you.

• High ownership: your judgment sets how the site runs, with a path to regional operations as FarmGPU opens more sites.

Compensation

• $155,000 – $190,000 base salary, plus equity and an on-call stipend
• Full-time, on-site in Los Angeles, CA.

Apply Now

About the company

FarmGPU

Actively Hiring More jobs Neocloud GPU compute with focus on storage11-50 Employees

San Francisco

11-50

Cloud Computing

Artificial Intelligence

Cloud Infrastructure

Blockchains

Learn more about FarmGPU

Founders

Jonmichael Hands CEO • 3 years

Folsom

View the team

More jobs at FarmGPU

• Onsite Technician (3 openings) — Los Angeles, CAIn office · Los Angeles · $105k – $145k→
• Senior Technical Marketing EngineerOnsite or remote · Rancho Cordova · +1 more · $120k – $200k→
• Network Automation EngineerIn office · Rancho Cordova · $120k – $200k→

View all jobs at FarmGPU

Similar Jobs