Staff Backend Engineer
Staff Backend Engineer
LiteLLM is the world’s most popular open-source AI Gateway, used by Adobe, Netflix, NASA, NVIDIA and much of the Fortune 500. Every LLM call these companies make runs through our code, so when we break, they break. We’re a team of about 20, and we’re hiring senior and staff backend engineers to own the core of the gateway and keep it fast and reliable as traffic grows.
About the role
You’ll own critical parts of the LiteLLM proxy: routing, auth, caching, spend tracking, guardrails and observability. The job is to make these systems correct, fast and dependable at 1B+ requests a month. That means hunting down silent failures, fixing the root cause in code rather than adding another dashboard, and refactoring components that keep breaking without changing how they behave. You’ll set the bar for how our mostly early-career team writes, tests and ships backend code.
What you’ll do
Own core proxy components end to end, from design through production, including the incidents
Find and remove whole classes of bugs, like race conditions, stale cache state across replicas, and errors that get silently swallowed instead of shown to users
Make sure every guardrail and policy check (e.g. apply_guardrail ) is logged, traceable and fails loudly when it should
Build guardrails that run inside the gateway and block common attacks on LLM APIs, MCP servers and agents
Keep our observability integrations (Datadog, Splunk, Prometheus, OpenTelemetry) accurate at 1B+ requests a month, with minimal latency overhead and no memory leaks (e.g. from Prometheus metric cardinality)
Help migrate performance-critical paths from Python to Rust without breaking behavior
Raise the bar for the team on code review, testing and debugging
What we’re looking for
5+ years building backend systems in production, with real ownership of something mission-critical: payments, gateways, core APIs, data stores, or anything else where downtime hurts customers
Strong Python, and experience with async or high-throughput services (asyncio, FastAPI or similar)
A track record of making production systems more reliable through code: refactors, re-architectures, migrations, or fixing a root cause for good
An excellent debugger across services, databases (Postgres, Redis) and distributed systems
High ownership, meaning you take responsibility for the customer impact of what you ship
BS/MS in Computer Science or a related field, or equivalent proof you’re exceptional
Nice to have
Rust, Go or C++ in production
Time at a fast-growing startup with a large, real codebase
Experience with LLM APIs, proxies or API gateways
Open-source contributions
Why LiteLLM
Your code runs in the critical path for some of the biggest companies in the world
Real ownership of core systems on a small, senior-leaning team
Profitable, growing fast, backed by Kleiner Perkins
$250-$350k base + equity, health, dental and vision. Onsite in San Francisco.
About LiteLLM
LiteLLM is a Python SDK and proxy server for calling 100+ LLM APIs in the OpenAI format, with 60K+ GitHub stars.
Source: the employer's careers page. Last checked 2026-11-20. Posted 2026-11-20.
More jobs at LiteLLM · Jobs in San Francisco · On-site jobs in United States · Software development jobs