About the company

One of our clients is a London based AI company building agent-based software that takes over complex back-office workflows for mid-sized financial and professional services firms. They are a small, senior team with their first customers already running in production, and they are now growing the engineering group to ship faster and take on larger accounts.

The role

They are looking for an AI Engineer to own LLM and agent features from prototype to production. You would work directly with the founders and the product lead, decide how agents plan, call tools and recover from errors, and see your work used by real customers within weeks, not quarters.

Responsibilities

• Build LLM powered product features and agentic workflows that solve real user or business problems.
• Integrate language models with APIs, databases, internal tools and external data sources.
• Develop retrieval augmented generation (RAG), tool calling, structured outputs and multistep workflows where appropriate.
• Improve quality and reliability through testing, evaluations, monitoring and iterative feedback.
• Deploy and maintain production services, balancing performance, security, scalability and cost.
• Work with product, design and customer facing teams to translate needs into practical technical solutions.

Core qualifications

• Strong software engineering fundamentals and hands-on experience with Python or TypeScript.
• Experience building applications or services that integrate LLM APIs or applied AI capabilities.
• Working knowledge of RAG, embeddings, prompt design, tool/function calling or agent workflows.
• Experience with APIs, databases, testing, cloud deployment and production debugging.
• Ability to communicate technical tradeoffs and work effectively in a fast moving environment.

Nice to have

• Experience with LangGraph, LangChain, LlamaIndex or similar frameworks.
• Experience with AI evaluations, observability, guardrails, model routing or inference optimization.
• Early-stage startup experience or ownership of an AI feature from prototype through production.