JOB-1428
An AI platform company runs a dedicated team for agents with long time horizons, which perform sustained, real-world work for customers. It is looking for senior and staff engineers to own how agent quality is measured and improved, and to advance platform reliability.
What you’ll own
- Set the strategy for evaluations and the practice of error analysis across the agent platform
- Design and build the agent knowledge systems and evaluation tooling behind the agents
- Lead investigations into agent failures and convert findings into systemic reliability fixes
- Strengthen the backend systems that execute, evaluate and monitor agents at scale
- Move the roadmap for dependability and quality forward with the wider AI group
- Coach engineers and raise the standard for eval rigor
About you
- Six to fifteen years as a software engineer, at senior, staff or tech lead level
- Prior end-to-end work on production AI agents or platforms, including agentic loops, evaluations and failure analysis
- Strong backend, data platform or full-stack background, and comfortable owning systems at the platform level
- Customer-facing agents used at scale, not only internal tools or simple chatbots
- A habit of finding the systemic cause of a failure and fixing it
- A multiplier mindset that lifts colleagues
- Distributed systems experience is a plus
Hybrid in the San Francisco Bay Area, roughly three days a week on-site. Visa transfers considered.
