AI Engineer

At Potpie, we are building infrastructure for the next generation of AI-native developer tools. We focus on enabling agents that can reason over real codebases and perform meaningful engineering tasks. Our platform transforms repositories into structured, graph-backed representations that agents can traverse, analyze, and act on reliably in production. We are looking for a Machine Learning Engineer to help design and ship production-grade LLM agent systems. You will work closely with the founding team on systems that combine large language models, knowledge graphs, and program analysis. The work spans agent architecture, context construction, and improving reliability of long-running workflows.

In this role, you will build and improve long-running agent systems that remain reliable over complex codebases, with a focus on mitigating context rot and developing specialized AI-native tooling for enterprise environments.

Key Responsibilities * Build and improve code-aware reasoning systems on top of knowledge graphs * Develop multi-step agent workflows for enterprise systems * Improve context retrieval and ranking from large codebases * Optimize latency, cost, and reliability of agent pipelines

Qualifications * Strong Python skills * Experience with multi-agent systems. Bonus point if it's open-source * Experience with the latest LLM based AI tooling * Strong debugging and systems thinking ability

Why join [potpie.ai]( * Work on production-grade agentic AI systems * Tackle deep problems in code understanding at scale * High ownership and direct impact * Collaborative, low-bureaucracy environment * Competitive salary and meaningful equity