Compute Fleet Reliability Engineer | GPU/TPU

Jobtailor; San Francisco (CA); On-site; USD 170,000 - 240,000; Full time

Job summary

FluidStack is seeking a hands-on reliability engineer to own the compute fleet health at scale in production environments, including Kubernetes and bare metal. You will build metrics pipelines, alerting, and a unified health view for GPUs/TPUs, and automate repair workflows from detection to service.

You will design the XPU qualification platform, define good-enough baselines, and own Redfish/BMC tooling and telemetry.

Qualifications

• Toil as a bug: manual steps in a repair workflow are a backlog item.
• Strong hardware fault analysis at firmware/silicon level.
• Walk into the fog, build the map, and explain it to others.
• Learn quickly in unfamiliar domains; this is valued over prior expertise.
• Carry a pager: respond to incidents, write postmortems, fix root causes.
• Fluent with AI tooling and AI coding environments.

Responsibilities

• Own compute fleet health end to end with metrics pipelines and alerts.
• Turn repair into automation from detection to service.
• Design and expand the XPU qualification platform for GPUs/TPUs.
• Own Redfish and BMC tooling and fleet telemetry.
• Own reliability, scalability, and operation of the compute fleet at scale.

Skills

Toil as a bug

Hardware intuition

Ambiguity navigation

Rapid learning

Pager/Incident response

AI tooling fluency

Tools

Redfish tooling

BMC tooling

IPMI tooling

Prometheus

Grafana

Go

Python

Job description

FluidStack is seeking a hands-on reliability engineer to own the compute fleet health at scale in production environments, including Kubernetes and bare metal. You will build metrics pipelines, alerting, and a unified health view for GPUs/TPUs, and automate repair workflows from detection to service.

You will design the XPU qualification platform, define good-enough baselines, and own Redfish/BMC tooling and telemetry.