About the Role
Build developer tools that help AI agent teams evaluate and improve agents using execution traces. This product engineering role combines hands-on development with community engagement, so insights from builders can directly shape what gets shipped.
What You'll Do
Build and ship features for agent evaluations, replay, cohorts, and developer experience.
Share technical work through posts, demos, and conversations on public developer channels, and bring useful feedback into product decisions.
Work across feature development, customer demos, marketing activities, and community engagement as priorities require.
Stay connected to AI agent builders and evaluation practitioners, including those using agent frameworks and custom runtimes.
Connect community feedback with product development, and communicate shipped improvements to the developer community.
What We're Looking For
At least 3 years of experience building or shipping production software, including production experience building or evaluating AI agents.
Production proficiency in Python and hands-on experience with agent frameworks such as LangGraph, Claude Code, or equivalent tools.
Experience with LLM or agent evaluation systems, trace data, observability, or debugging agent execution.
Open-source contributions or developer tools with demonstrated use by other developers.
Experience shipping SDKs, APIs, CLI tools, frameworks, or libraries, with attention to documentation, ergonomics, and demos.
Experience supporting distribution or go-to-market for products you helped engineer, and communicating technical work publicly.
Comfort working across responsibilities in an early-stage or otherwise high-uncertainty environment.
Compensation & Benefits
Base salary: $110,000 to $190,000 USD annually. Visa sponsorship is available.
Location
Remote, with the primary location in Munich, Bavaria, Germany.