Provider Ops SRE: Reliability, Failover & Automation

OpenRouter, Inc; United States; Vor Ort; USD 150.000 - 210.000; Vollzeit

Zusammenfassung

OpenRouter, Inc. is hiring an AI Inference SRE to own the operational health of our provider supply. You’ll build monitoring, run on-call, and improve failover across 80+ providers and thousands of endpoints.

You’ll join the Provider Operations team and report to the Provider Operations Manager. Candidates should have strong observability experience and a product mindset for reliability in high-traffic AI workloads.

Qualifikationen

• 4+ years in SRE or production engineering.
• Observability with metrics, tracing, logs and SLOs.
• Software engineer who writes tools (TypeScript/Python).
• Experience with distributed systems failure modes.
• Calm incident commander who communicates under pressure.
• Understanding of how LLM inference is served.

Aufgaben

• Provider health and observability: monitor latency, throughput, error rates, uptime, and output correctness.
• Detection and failover: quick degradation detection and traffic rerouting.
• Incident response: on-call, triage, mitigations, postmortems.
• Provider accountability: telemetry to scorecards and escalation point.
• Quality regression detection: canaries and evals for regressions.
• Automate the toil: replace manual provider-ops work with tooling.
• Capacity and launch readiness: load-test endpoints before big launches.

Kenntnisse

Observability tooling

SRE / Production engineering

TypeScript

Python

Distributed systems

Incident response

LLM inference

Tools

Postgres

ClickHouse

GCP

Cloudflare Workers

Vercel

Jobbeschreibung

OpenRouter, Inc. is hiring an AI Inference SRE to own the operational health of our provider supply. You’ll build monitoring, run on-call, and improve failover across 80+ providers and thousands of endpoints.

You’ll join the Provider Operations team and report to the Provider Operations Manager. Candidates should have strong observability experience and a product mindset for reliability in high-traffic AI workloads.