Applied AI Engineer

About the Role

You will own the agentic infrastructure at a well-funded AI infrastructure startup, building the systems that let AI agents execute reliably, retain context over months, and know when to involve a human. This is a founding engineer role where your decisions directly shape how agents plan, remember, fail, and recover in production.

What You'll Do

• Build and maintain core infrastructure that enables agents to execute tasks reliably across dozens of iterations.
• Design and implement memory systems that retain months of client context extracted from messy, real-world operational data.
• Develop eval harnesses that teams actually trust to make production deployment decisions.
• Own the full loop: build, measure, break, fix, and improve agent systems based on production feedback.
• Ship agent systems directly to real customers and iterate on them based on live feedback.
• Contribute across infrastructure, orchestration, customer collaboration, and early hiring as needed.

What We're Looking For

• 3 or more years shipping LLM agents that ran unattended in production for real users, with a clear understanding that the model is the easy part.
• Proven experience building eval harnesses that were actually used to make production shipping decisions.
• Strong Python proficiency applied to production systems, not prototypes.
• Experience with agent memory management, context retrieval, and orchestration on real operational data (not demo RAG pipelines).
• Hands-on experience with agent orchestration frameworks, workflow management, or infrastructure tooling.
• Experience implementing failure recovery, retry logic, or reliability patterns in production agent systems.
• Bonus: open-source agent tooling contributions, experience with human-in-the-loop or approval workflows, or a founded/led technical project.
• Exceptional intensity and execution, comfortable owning work well beyond a narrow specialty at an early-stage company.