Site Reliability Engineer, Compute

2 months ago

Austin, TX, USA +3 moreSenior

Base Salary

$208k - $269k/yr

Responsibilities

Own end-to-end health, reliability, scalability, and operations for a large GPU compute fleet across Kubernetes and bare-metal environments.
Build metrics pipelines, alerting, observability, and unified fleet-health views.
Automate compute failure workflows from detection and triage through parts management, RMA, and return to service.
Design and expand GPU qualification platforms for burn-in, performance baselining, and new product introduction.
Own Redfish and BMC tooling, including firmware-level telemetry, log collection, and low-level fleet access.
Run incidents, write postmortems, and address systemic reliability causes.
Build production automation and eliminate manual repair and operational toil.

Experience shipping production automation that other teams depend on, using any programming language and AI coding tools.
Comfort reasoning about hardware, firmware, silicon-level failure modes, and software infrastructure.
Ability to operate independently in ambiguous, high-urgency environments and rapidly learn unfamiliar domains.
Willingness to participate in pager-based incident response and own incidents through systemic resolution.
Fluency with AI tooling such as LLM APIs, MCP servers, agentic frameworks, Claude Code, Cursor, or similar tools.
Hardware lifecycle management and RMA automation experience is preferred.
Experience with BMC/Redfish or IPMI tooling is preferred.
Experience with GPU qualification or burn-in frameworks is preferred.
Experience with workflow and orchestration engines such as Temporal or Cadence is preferred.
Experience with metrics and alerting tools such as Prometheus or Grafana is preferred.
Go or Python experience is preferred.

201-500 employees