AI Engineer (LLM, RAG, Agents)

Role overview

Build and operate production AI systems that use large language models, retrieval-augmented generation, and autonomous agents. The work applies these techniques to practical workflows such as document processing, sales automation, and knowledge-grounded customer support, with a focus on measurable accuracy, operating cost, and business impact.

Responsibilities

- Develop production LLM integrations, prompts, fine-tuning approaches, and evaluation methods. - Design RAG pipelines for document ingestion, chunking, embeddings, indexing, hybrid retrieval, and reranking. - Build agents with tool use, multi-agent coordination, and human review checkpoints. - Create evaluation pipelines and use their results to improve system quality. - Deploy guardrails, tracing and observability, and cost controls. - Work with client technical leads on architecture decisions and keep current with changes in models and tools.

Requirements

- Four or more years building production AI or machine-learning systems. - Strong Python skills, including FastAPI, asynchronous programming, and typing; plus TypeScript or Go for service development. - Practical experience with OpenAI and Anthropic models and at least one open-source model such as Llama or Mistral. - Production RAG experience, including chunking, embedding selection, hybrid search, and reranking. - Production experience with at least one vector database, such as Pinecone, Weaviate, Qdrant, or pgvector. - Familiarity with LangChain, LlamaIndex, or LangGraph, plus cloud deployment on AWS or GCP and Docker. - Business-level English for technical writing, code review, and daily client discussions.