AI Engineer

Most companies try to meet expectations, dunnhumby exists to defy them. Using big data, deep expertise and AI-driven platforms to decode the 21^st century human experience - then redefine it in meaningful and surprising ways that put customers first. Across digital, mobile and retail. For brands like Tesco, Coca-Cola, Procter & Gamble and PepsiCo.

We are looking for a hands-on AI Engineer to build production-grade AI systems powered by LLMs and agents. This role goes beyond prototypes, you'll design reliable, observable, and safe AI applications at scale.

Key Responsibilities

• Design and develop LLM-powered applications and AI agents for real-world use cases
• Build and orchestrate agent workflows using LangChain and LangGraph
• Implement human-in-the-loop (HITL) systems for controlled AI decision-making
• Design and optimize RAG pipelines
• Advanced retrieval strategies (hybrid search, re-ranking, filtering)
• Effective chunking strategies (semantic chunking, hierarchical chunking, context-aware splitting)
• Integrate and work with MCP (Model Context Protocol) servers for tool and context orchestration
• Integrate with Vertex AI and/or Azure AI Foundry for model deployment and lifecycle management
• Implement guardrails (prompt injection defense, output validation, policy enforcement)
• Build evaluation frameworks for model quality, hallucination detection, and regression testing
• Ensure observability (traces, metrics, logs) for LLM applications in production
• Collaborate with platform, data, and product teams to deliver end-to-end solutions

Required Skills

• 3-5 years of experience in AI/ML or backend engineering
• Strong programming skills in Python
• Hands-on experience with LLMs, NLP, and AI agents
• Strong experience with LangChain and LangGraph
• Experience implementing human-in-the-loop workflows
• Solid understanding of RAG systems, including retrieval and chunking strategies
• Experience working with or understanding MCP servers / tool orchestration frameworks
• Strong understanding of evaluation frameworks (LLM evals, prompt testing, benchmarking)
• Experience designing and implementing AI guardrails and safety mechanisms
• Proficiency in cloud platforms (GCP or Azure)
• Experience with Vertex AI and/or Azure AI Foundry is required
• Experience with APIs, microservices, and scalable system design

Good to Have

• Experience with vector databases (Pinecone, Weaviate, FAISS, etc.)
• Familiarity with observability tools for AI systems (tracing, token usage, latency tracking)
• Experience with Kubernetes / containerized deployments
• Exposure to event-driven systems (Kafka, Pub/Sub, etc.)
• Knowledge of multi-agent architectures and planning/reasoning patterns