Keep our agents healthy in production. You'll build the monitoring, tracing, and eval pipelines that catch regressions and keep cost, latency, and behavior in bounds.

What you'll own: - Tracing and observability for tool-using agents - Automated eval pipelines wired into CI - Cost and latency optimization, plus incident response

Ideal background: - SRE/MLOps experience applied to ML or LLM systems - Fluency with metrics, tracing, and alerting stacks - A reliability mindset for non-deterministic systems