Artificial Intelligence Engineer

Overview

The Artificial Intelligence Engineer will work on developing and deploying production-grade applications featuring large language models (LLMs). The role involves collaborating with a team to create RAG pipelines and integrate various AI tools, with a strong focus on latency optimization and cost efficiency. The contractor will have opportunities for hybrid and remote work within the UK, with the project set to last between three to six months with potential extensions.

Responsibilities

  • Build and implement production RAG pipelines and AI agents.
  • Integrate commercial and open-source LLMs into applications.
  • Evaluate and tune LLMs for optimal performance in terms of latency and cost.
  • Develop APIs using Python, preferably with FastAPI or similar frameworks.
  • Design agent and tool-calling mechanisms, including MCP or function-calling patterns.
  • Oversee LLM evaluation and observability through tools such as RAGAS and DeepEval.
  • Serve and deploy open-source models using technologies like vLLM or TGI.

Requirements

  • Demonstrated commercial experience in delivering LLM applications to production.
  • Proficient in Python, with experience in FastAPI or equivalent for API development.
  • Familiarity with LLM APIs such as OpenAI, Anthropic, Gemini, or Azure OpenAI.
  • Experience with frameworks like LangChain, LangGraph, or LlamaIndex.
  • Knowledge of vector databases including Pinecone, Weaviate, or Elasticsearch.
  • Understanding of RAG design concepts, including chunking strategies and re-ranking.
  • Experience with open-source model serving solutions and cloud deployment on AWS, GCP, or Azure.