LLM Engineer

Our clients want LLM systems that answer correctly from their own data and can be trusted in front of their staff and customers. You’ll build RAG pipelines, AI agents and copilots that connect large language models to client documents, databases and tools. Your job is to make them accurate, measurable and affordable to run, well past a demo that works once.

• Design and build RAG pipelines, from document chunking and embeddings to retrieval, re-ranking and answer generation
• Build agents that call tools and APIs to complete multi-step tasks for users
• Create evaluation sets and automated tests that measure answer accuracy and catch hallucinations
• Choose between hosted models (OpenAI, Anthropic, Gemini) and open-source models based on cost, latency and data privacy
• Ship production APIs with FastAPI, including logging, caching and cost tracking

• 2+ years of software engineering, with at least 1 year building LLM applications that real users rely on
• Strong Python and solid API design skills
• Hands-on experience with a vector database such as Pinecone, Weaviate, Qdrant or pgvector
• Experience with an orchestration framework such as LangChain, LlamaIndex or LangGraph, and the judgement to know when not to use one
• Clear understanding of prompt design, context windows, token costs and model limits

• Experience fine-tuning open-source models with LoRA or QLoRA
• Experience serving models with vLLM or Ollama
• Familiarity with LLM evaluation tools such as Ragas, DeepEval or LangSmith
• Experience building for regulated industries such as healthcare or pharma