Site Reliability Engineer, Provider Operations

present of AI; United States; Remote; USD 120,000 - 180,000; Full time

Job summary

OpenRouter seeks a Site Reliability Engineer for Provider Operations to keep a broad set of AI model providers healthy and traffic flowing. You will design observability, build tooling in TypeScript or Python, and lead incident response with provider partners to minimize outages.

The role emphasizes cross-team collaboration, strong monitoring, and the ability to test and validate provider APIs to ensure performance and reliability across streaming, calls and cache interactions.

Qualifications

• Solid reliability engineering experience.
• Strong monitoring and alerting skills.
• Ability to write tooling in TypeScript or Python.
• Calm under outages and clear collaboration with provider teams.

Skills

Observability

SLO design

Incident response

TypeScript tooling

Python tooling

Tools

TypeScript

Python

Job description

Site Reliability Engineer, Provider Operations

The role is about keeping a large set of AI model providers healthy so customer traffic keeps flowing. It asks for solid reliability engineering experience, strong monitoring and alerting skills, and the ability to write tools in TypeScript or Python instead of repeating manual steps. A strong candidate stays calm during outages and can work clearly with outside provider teams.

How it fits the AI era

AI changes this work because the systems being watched are language models, so problems can be quiet, like a model that gets worse or a tool call that stops working, rather than a simple outage. Automated checks and evals take over much of the detection and repetitive upkeep. Judgment during incidents, trust with provider partners, and deciding what counts as acceptable quality stay human.

How it builds AI experience

This role gives close, daily exposure to how many different model providers serve inference, including streaming, tool calling, caching, and speed tradeoffs. Someone preparing could build a small tool that tests several model APIs for speed and answer quality, then track the results over time. Reading provider documentation and comparing how their behavior differs is also useful practice.

Worth showing

• Observability and SLO design
• TypeScript or Python tooling
• Incident response leadership
• Synthetic monitoring or evals

This is our own read, written from the employer's posting and the company's Spotlight. The employer's posting is the source of truth for the role.

About OpenRouter

OpenRouter is a unified API gateway and marketplace that routes requests across hundreds of AI models from many providers, per the company.