Staff Software Engineer

Moveworks is a platform that enables AI-powered automation of employee workflows across enterprise systems. This posting describes a role focused on building the judgement layer of the agent evaluation platform, including rubrics, judges, calibration against human labels, and the methodology to turn evaluation scores into actionable learning signals. The role covers three areas: Eval orchestration at scale, Agent observability and tracing, and Stateful simulation. You will work on translating complex agent trajectories and outcomes into reliable measurements, enabling agents to learn from their mistakes and improve performance. Responsibilities include implementing end-to-end runtime environments for multi-turn agent scenarios, designing and operating large-scale evaluation harnesses, and developing observable tracing of prompts, tool calls, and plan updates. You will contribute to OpenTelemetry-native observability, ensure traceability across async boundaries, and maintain hermetic, repeatable per-run simulations with realistic enterprise-system mocks. Requirements include 8+ years building production backend or infrastructure, strong proficiency in Python or Go (ideally both), and experience designing systems that handle real traffic at scale. You should be comfortable turning non-deterministic behavior into measurable signals and navigating novel, open research problems around LLMs judging LLMs. This role requires in-office presence in Mountain View, CA or equivalent, with flexibility and collaboration across teams. Apply through the ServiceNow careers portal.