Backend Engineer, Agent Platform (Vertical SaaS AI)

A vertical software company is building a large-scale platform that puts autonomous AI agents to work inside an industry, and its agent platform team needs a backend engineer at the mid or senior level. The system has to stay dependable at scale, so the team wants someone who cares about product impact as much as system design.

What you’ll own

Runtime services for agents, covering orchestration, tool invocation, retries, idempotency, rate limits and state handling

Dependable, versioned APIs for internal and external consumers, plus webhook and event hooks and adapters for partner systems

Data models that hold agent state, past workflow runs, access rules and tenant separation

Integrations such as OAuth flows, sync engines and connectors, built with good monitoring and graceful handling of failures

Throughput, latency, queues, caching and inference spend, plus guardrails against runaway tasks

Reliability practice (SLIs and SLOs, distributed tracing, timeouts, circuit breakers, incident response) and higher engineering standards through testing, runbooks, postmortems and mentoring

What you bring

Three or more years of backend engineering, with exposure to AI or conversational agents

Strong distributed-systems fundamentals: concurrency, queues and workers, caching, production operations

Recent hands-on experience shipping LLM agents with tool calling and multi-step orchestration, improved through evals, prompt iteration or fine-tuning

Experience scaling production AI/ML platforms in a fast-paced startup

Daily use of AI coding assistants, and the ability to explain how they speed up your work

Comfort with ambiguity and a proactive, autonomous working style

Hybrid in the South Bay Area, three days a week in the office. Visa transfers are welcome.