Staff Software Engineer – Product Quality & Release Automation

LiteLLM; San Francisco (CA); On-site; USD 180,000 - 240,000; Full time

Job summary

LiteLLM is the world's most popular AI Gateway, trusted by top companies. We are seeking a Staff Software Engineer to own product quality across testing, CI, and release systems at scale.

You will own LiteLLM's quality across 100+ provider integrations and a high-throughput flow from contributions to production, building conformance tests and release tooling to keep speed and stability aligned.

Qualifications

• Experience at a startup (Seed to Series D) with a willingness to wear multiple hats.
• Built test or release infrastructure for a system with a large external integration surface, or maintained CI for an OSS project with heavy outside contribution.
• Strong debugging depth: can enter an unfamiliar codebase and find real root causes, not just symptoms.
• Opinions about what makes a test suite worth trusting.
• Track record of raising quality on a team without becoming the bottleneck.
• Open source contributions on GitHub or similar.

Responsibilities

• Own product quality end to end and drive regressions toward zero.
• Build a provider conformance test suite covering streaming, tool calling, structured output, error mapping, cost tracking, and auth modes.
• Own CI and release tooling that is fast, deterministic, and trusted enough that a red gate blocks a ship.
• Build canary and synthetic traffic detection to catch upstream provider drift before customers do.
• Own release channels so we ship daily at the edge while enterprise stays stable.
• Do real root cause analysis and turn every escaped defect into a permanent test.
• Raise the engineering bar through code and review, not through process documents.

Job description

LiteLLM is the world's most popular AI Gateway, trusted by top companies. We are seeking a Staff Software Engineer to own product quality across testing, CI, and release systems at scale.

You will own LiteLLM's quality across 100+ provider integrations and a high-throughput flow from contributions to production, building conformance tests and release tooling to keep speed and stability aligned.