Site Reliability Engineer, Compute

About the Company

Fluidstack is building civilization-scale infrastructure for AI, focused on delivering hundreds of gigawatts of compute by designing, building, and operating massive data centers.

Responsibilities

• Own compute fleet health end to end, including metrics pipelines, alerting, and unified health views.
• Automate deployment and repair pipelines to handle GPU failures from detection through return to service.
• Design and expand the GPU qualification platform for burn-in and performance baselining.
• Manage Redfish and BMC tooling for firmware-level telemetry and low-level access.
• Ensure end-to-end reliability and scalability of the global GPU fleet through aggressive automation.

Requirements

• Experience shipping production automation that other teams depend on.
• Strong instinct for hardware and ability to reason about failure modes at the firmware and silicon level.
• Proficiency with AI coding tools (e.g., Claude Code, Cursor) and AI tooling/frameworks.
• Ability to manage incidents, perform postmortems, and fix systemic causes.
• Comfort working in ambiguous environments and learning unfamiliar domains quickly.

Preferred Qualifications

• Experience with hardware lifecycle management and RMA automation.
• Expertise in BMC/Redfish or IPMI tooling.
• Experience with GPU qualification or burn-in frameworks.
• Knowledge of workflow orchestration engines like Temporal or Cadence.
• Proficiency with Prometheus, Grafana, Go, or Python.