GPU Cluster Infrastructure Engineer High-Perf

Beam; San Francisco (CA); Hybride; USD 166 000 - 277 000; Temps partiel

Avantages offerts par ce poste

Competitive salary and meaningful -equ

Health, dental, and vision benefits

Opportunities to participate in events

Fitness stipend

Learning budget

Résumé du poste

Beam is building out its own GPU capacity and seeks an experienced contractor to stand up high-performance GPU clusters, from design review through bring-in.

You will review designs, validate cabling and InfiniBand fabric, perform burn-in, and coordinate with vendors on storage and compute. The role requires hands-on data center work with remote hands capability and strong documentation skills.

Qualifications

• Experience building and operating NVIDIA HGX/DGX clusters in production.
• Hands-on InfiniBand subnet management and fabric bring-up.
• Experience with firmware, BMC/Redfish, DCGM, NCCL testing, PXE and imaging.
• Experience with parallel storage (WEKA, VAST, GPFS, Lustre).
• Able to work both on the data center floor and remotely.
• Strong troubleshooting and documentation skills.
• Bonus: NVIDIA platforms, bare-metal cloud operations, Ansible, Prometheus/Grafana, NVIDIA certifications.

Responsabilités

• Review cluster designs and BOMs across compute, networking, and storage.
• Lead acceptance testing: cabling and optics, InfiniBand fabric bring-up, burn-in.
• Stand up and validate high-performance storage with vendor teams.
• Build the out-of-band management layer and firmware baselines; secure management plane.
• Integrate telemetry into observability stack with alerts and health checks.
• Write runbooks, as-builts, and remote-hands procedures.
• Provide escalation support after go-live and help ramp up.

Connaissances

GPU cluster operation

Hardware troubleshooting

Documentation & runbooks

Vendor coordination

Remote hands coordination

Clear technical communication

Telemetry and observability

Outils

NVIDIA HGX/DGX

InfiniBand subnet management

BMC/Redfish

DCGM; NCCL testing

PXE/imaging

WEKA

VAST

GPFS/Lustre

Prometheus/Grafana

Ansible automation

Description du poste

Beam is building out its own GPU capacity and seeks an experienced contractor to stand up high-performance GPU clusters, from design review through bring-in.

You will review designs, validate cabling and InfiniBand fabric, perform burn-in, and coordinate with vendors on storage and compute. The role requires hands-on data center work with remote hands capability and strong documentation skills.