Founding Harness Engineer
Core
Design and build the agent runtime and evaluation systems for an AI coworker to act reliably in real team workflows.
Role type
Founding IC machine-learning engineer (agent runtime & evaluation)
Builds
Agent runtime, evaluation systems, and regression tests for tool-using LLM agents
Domain
Artificial Intelligence / LLM Agents / Software Engineering
Deliverable
production ML models
Required skills
LLM agent development, system prompt design, evaluation and regression testing, tool use implementation, approval workflow design, multi-tenant data separation
Preferred skills
Experience shipping tool-using LLM agents beyond demos, building evaluation harnesses for LLM behavior, understanding safe tool execution and multi-user data risks
Technologies
LLMs, agent frameworks, testing frameworks
Responsibilities
Design and build the agent runtime for context, tool selection, and approval requests; Write and refine system prompts to balance proactivity with safety; Build evaluation and regression tests for real tasks and failure cases; Improve reliability for long-running tasks and retries; Implement safeguards like approval gates and data separation; Work with founding team to take prototypes to production
