Software Engineer, Agentic AI Systems

We are looking for someone with deep, genuine knowledge of how agentic AI systems actually work, not just familiarity with the frameworks. Someone who has thought hard about why these systems fail, what makes them reliable, and how to measure their behavior in production.

You will own:

• Design and architecture of agent orchestration pipelines that handle real-world non-determinism gracefully
• Memory and state management across long-horizon, multi-step reasoning tasks
• Tool-use reliability: schema design, error propagation, fallback strategies across external APIs
• Evaluation frameworks for agent behavior, because intuition alone cannot scale
• Inference optimization: model routing, latency management, cost tradeoffs at production scale

What We Are Looking For

• Deep knowledge of agentic frameworks in real systems: LangGraph, AutoGen, CrewAI, or equivalent
• Strong Python fundamentals: async patterns, typed codebases, structured testing
• Genuine understanding of multi-agent failure modes and how to design around them
• Ability to evaluate model behavior systematically and build observability into AI systems from day one
• Comfort operating with ambiguity and writing specs when none exist
• US Citizenship required

What This Is Not

• We want to be honest with you about what this role is and is not. This is not a role where you integrate third-party APIs and call it AI engineering. The problems here are structurally hard and largely unsolved. If you are looking for a well-defined roadmap and clear requirements handed to you, this is probably not the right fit.
• If you find open problems energizing and you have built systems that broke in interesting ways and taught you something real, we want to talk.

What You Get

• Direct influence on technical direction from day one
• Real infrastructure and compute budget
• Close collaboration with founders on product and architecture

Pay: From $95,000.00 per year

Work Location: In person