Python AI Engineer

Overview of the role

The Python AI Engineer will build the agent and retrieval layer of an enterprise AI assistant serving thousands of employees, where an orchestrator invokes skills and routes to specialized agents exposed as internal services. This is a production system that already exists and keeps evolving, not a greenfield prototype, with ownership over evaluation coverage for the behavior shipped.

Key responsibilities

- Building agent and retrieval capabilities for a production AI assistant.

- Building LangGraph flows and RAG pipelines over structured and unstructured sources.

- Holding evaluation coverage on the AI behavior shipped.

- Extending the existing Next.js front end where features require it.

Requirements

- Python: 5+ years of hands-on experience, 8+ years for lead-level scope.

- Production LLM Systems: Proven track record building, maintaining, and evolving systems, not just taking them from zero to one.

- LangChain/LangGraph: Hands-on experience with these frameworks.

- RAG: Experience over structured and unstructured sources.

- AWS: Working experience with the platform.

- Availability: Must be available to work PST hours.

- Language Proficiency: Advanced level of English.

Additional skills preferred

- LLM Evaluation: Experience with tracing tools such as LangSmith, LLM-as-judge, or golden datasets.

- React/TypeScript: Some background with these technologies.

- Multi-Agent Orchestration: Working experience.

- Related Technologies: Familiarity with vector databases, MCP servers, or streaming agent interactions.