Senior Agentic AI Engineer

We’re looking for a Senior Agentic AI Engineer to join our team in Leeds City Centre on a permanent basis. This is a hybrid role, with two days in the office and three days working remotely each week. This role sits within a newly established AI engineering squad, working across the full agentic stack. You’ll design agent architectures and RAG pipelines, and deliver autonomous agents into production to replace manual processes. It’s a hands‑on, delivery‑focused role with immediate impact from day one. Agentic AI engineering is a rapidly evolving field, so we don’t expect years of experience in a discipline that’s still emerging. What matters is a track record of building real, production‑grade solutions, strong software engineering fundamentals, and the curiosity to stay ahead. If you’ve been working in this space for 18+ months and can demonstrate live systems you’ve built, we’d love to hear from you.

What will you be doing?

Design and build multi‑agent systems for autonomous reasoning, planning, and task execution
Architect orchestration patterns including tool use, memory, reflection loops, and human‑in‑the‑loop handoffs
Deliver solutions across use cases such as developer productivity, knowledge retrieval, and process automation
Leverage GCP‑native and enterprise‑grade managed services ahead of building custom infrastructure
Develop end‑to‑end RAG architectures (ingestion, chunking, embedding, retrieval, reranking)
Select and manage appropriate vector databases and build pipelines to turn unstructured data into usable insights
Apply knowledge graph approaches (e.g. Neo4j) where deeper relational or semantic reasoning is needed
Define and implement evaluation frameworks, including automated testing and LLM‑based assessment
Monitor and optimise agents for cost, latency, quality, and production performance
Collaborate across engineering and business teams to build scalable, compliant (GDPR/EU AI Act) and production‑ready agentic solutions

Essential

Proven experience building and deploying agentic AI systems in production (not just prototypes)
Strong Python engineering skills, with a focus on clean, tested, maintainable code
Hands‑on experience designing RAG pipelines (chunking, embeddings, retrieval, evaluation)
Experience working with at least one vector database in a production setting
Practical experience with LLM evaluation techniques (e.g. automated evals, LLM‑as‑a‑Judge, or similar)
Familiarity with agent frameworks (e.g. LangChain, LangGraph, CrewAI, AutoGen), with good judgement on when to use them
Solid understanding of LLM fundamentals, including prompting, tool use, structured outputs, and context management
Experience working with cloud platforms (GCP preferred; AWS or Azure also considered)

Desirable

Experience with Google Agent Development Kit (ADK) or Vertex AI Agent Builder
Understanding of MCP (Model Context Protocol) and agent‑to‑agent communication patterns
Experience with knowledge graphs or graph databases (e.g. Neo4j, JanusGraph)
Familiarity with LLMOps and observability tools (e.g. LangSmith, Langfuse)
Experience building solutions in regulated or enterprise environments
Background in BPO, AP automation, or document processing workflows

What’s in it for you?

Competitive salary + bonus
25 days annual leave with the option to purchase 5 more
Enjoy your birthday off, plus receive an additional 3 days of annual leave once you reach 5 years of service
Access to wellbeing and mental health benefits such as the Calm app, personal medical, critical illness cover and dental insurance, plus many more
Matched pension contribution up to 10%
Access to our car benefit scheme
Access to our online learning platform to continue to develop and grow your career with us
The chance to join an innovative, fast‑paced and passionate team

Liberty Global is an equal opportunity employer, committed to an inclusive environment and accommodating all candidates. We’re eager to hear from you, no matter your background.

#J-18808-Ljbffr

Full job description from WhatJobs. Apply for this role →