We are hiring an AI Engineer to build production-grade AI features inside our enterprise products. This is not a research role. You will ship working systems: retrieval pipelines, agentic workflows, structured extraction, and LLM-backed services that real users depend on every day.

You will work on features end to end, from problem framing and prototype through deployment, evaluation, and iteration.

What you will do

• Build and maintain RAG pipelines, agentic systems, and LLM-powered services using Python, FastAPI, and modern vector stores (Pinecone, Chroma, pgvector, or similar).
• Design prompt strategies, evaluation harnesses, and guardrails for production reliability. Track quality with real metrics, not vibes.
• Integrate LLM APIs (OpenAI, Anthropic, open-source via Hugging Face or local inference) into existing product surfaces. Decide when to use which model based on cost, latency, and accuracy tradeoffs.
• Ship features behind feature flags, instrument them, and iterate based on user signal and evaluation results.
• Collaborate with backend engineers on data pipelines, with frontend engineers on integration, and with product on scoping. Communicate clearly in writing and in standups.
• Document what you build so that the next engineer can extend it without asking you.

What we are looking for

• 3 to 5 years of professional software engineering experience, with at least one year hands-on with LLMs or applied ML in a production setting.
• Strong Python skills. Comfort with FastAPI or a comparable framework, async patterns, and clean API design.
• Practical experience with at least one of: LangChain, LlamaIndex, LangGraph, CrewAI, or a custom agentic framework you can defend. We care more about depth than the specific tool.
• Working knowledge of vector databases, embedding models, chunking strategies, and the failure modes of retrieval (hallucination, stale context, retrieval-recall tradeoffs).
• SQL competence. PostgreSQL preferred. You can write a window function without searching for it.
• Familiarity with Docker, Git, and at least one cloud platform (AWS, GCP, or Azure).

Nice to have

• Experience with healthcare data (HIPAA awareness).
• Fine-tuning or LoRA experience on open-source models.
• Background in evaluations: building benchmark sets, scoring frameworks, regression testing for LLM behavior.
• Experience with voice or multimodal pipelines.
• Open-source contributions or a public portfolio of agentic projects.