GPU Compute Ops Engineer - 24/7 Datahall Reliability

SpaceX Super Intelligence; Memphis (TN); On-site; USD 120,000 - 180,000; Full time

Job summary

SpaceXSI is building a 24/7 team responsible for 100% GPU uptime across compute and facility. You will own monitoring, incident response, and deployment testing for new data halls, while driving automation and standard operating procedures to improve reliability.

The role requires hands-on experience with data center telemetry, AI/ML infrastructure, and cross-functional collaboration across compute, network, and facility teams.

Qualifications

• Bachelor's degree in Mechanical, Electrical, or Computer Engineering or related scientific discipline.
• 2+ years in data center, critical facilities, power plant, or large-scale compute operations; or 6+ years equivalent experience.

Responsibilities

• Own monitoring architecture, alerting across GPU and host telemetry (DCGM, Prometheus/Grafana) and facility telemetry.
• Be part of 24/7 incident response, mitigation, and control authority from customer impact to facility.
• Own day-to-day site sustainment operations, resolving issues impacting rack-level compute and facility infrastructure.
• Disposition daily issues across Site Operations and Facility Operations.
• Lead testing campaigns to bring new data halls online and deploy production automation for datahall bring-up.

Skills

Prometheus/Grafana

NVIDIA DCGM

SCADA

AI/ML infra

NCCL

InfiniBand/RoCE

Slurm/Kubernetes

Redfish/IPMI

Python

C/C++

Incident command

Education

Bachelor's degree in Mechanical, Electrical, or Computer Engineering

Tools

Grok stack

Job description

SpaceXSI is building a 24/7 team responsible for 100% GPU uptime across compute and facility. You will own monitoring, incident response, and deployment testing for new data halls, while driving automation and standard operating procedures to improve reliability.

The role requires hands-on experience with data center telemetry, AI/ML infrastructure, and cross-functional collaboration across compute, network, and facility teams.