Founding Engineer for AI Agents

Job Summary

To support an innovative AI startup, the remote Founding Harness Engineer for AI Agents will build the agent runtime, manage system prompts, and develop evaluation and regression harnesses while collaborating closely with the founder and early users.

Key responsibilities

• Build the agent runtime, including context management and tool selection
• Write and tune system prompts, ensuring effective communication without spamming
• Own the evaluation and regression harness, testing prompt changes as code

Required qualifications

• Experience shipping tool-using LLM agents to real users in production
• Strong instincts for agent initiative, supported by test results
• Hands-on experience in building evaluation and regression suites for LLM behavior
• Solid backend skills in Python or TypeScript, including async jobs and state management
• Ability to navigate fuzzy problems from prototype to production and explain tradeoffs effectively