Senior AI Engineer

You'll build the platform's AI layer: turning the data that sits in the data lake into context that agents can actually use. It's end-to-end AI engineering — from retrieval (RAG: parsing, chunking, embedding) to building tools, prompt design, reasoning orchestration, and systematic evaluation of what works in production.

You decide architecture, pick the right models and techniques for each kind of data, and own the quality of the context that reaches the agent. It's one of the most central roles in the product: the bridge between raw data and the intelligence the customer sees.

Required:

• Python
• Solid understanding of LLMs — prompting, function calling/tools, reasoning, and how to evaluate output
• RAG pipelines, with judgment to choose a parser, parameterize chunking, and measure embeddings against the data context
• Vector databases (Pinecone, Weaviate, FAISS, Qdrant)
• Some agent framework (LangGraph, Pydantic AI, Agno, LangChain)

Nice to have:

• Knowledge Graph (Neo4j) and GraphRAG
• LLM fine-tuning
• MCP protocol