Quant / Agentic AI Developer - AVP

About the role

Barclays Markets is hiring a Quant / Agentic AI Developer (Assistant Vice President) in London to build a new Agentic AI capability in Global Markets. Working with traders, quantitative researchers, ML engineers and data engineers, you will design and deliver production-grade AI agents that automate and augment complex business workflows, reasoning over enterprise knowledge and interacting securely with tools and systems. The role is hybrid.

Key responsibilities

• Build production-grade AI agents that automate and augment complex business workflows, from prototype through to production.
• Help make existing technology and data platforms AI-ready, develop knowledge and retrieval capabilities and ensure solutions are secure, observable, cost-efficient and governed.
• Develop and implement quantitative models and strategies to derive insight into market trends and optimise trading decisions, pricing and risk management.
• Work with sales teams to identify client needs and develop customised solutions.
• Research, analyse data and run statistical modelling to understand market trends, pricing and risk dynamics.
• Maintain analytical libraries and provide expertise on quantitative methodologies and industry best practices.

Required profile

• Master's degree in a STEM field (or equivalent industry experience).
• Proven experience taking at least one LLM or Agentic AI application from prototype into production.
• Ability to work effectively with technical and business stakeholders.

Required skills

• Python, ML and software engineering, including testing, CI/CD, Git, containerisation and cloud-based AI workloads.
• LLMs and NLP, including prompt/context engineering, embeddings, structured outputs, retrieval and evaluation.
• RAG and enterprise knowledge solutions: document ingestion, chunking, vector databases, hybrid search, re-ranking, metadata and access-controlled retrieval.
• Agent frameworks such as LangGraph, Claude Agent SDK, OpenAI Agents SDK, Strands/Bedrock AgentCore or Microsoft Agent Framework.
• Highly valued: multi-agent architectures, MCP/tool integrations, LLM evaluation and guardrails, AWS AI services (Amazon Bedrock, SageMaker, Lambda, EKS, S3), AI-assisted DevOps, graph knowledge bases, model fine-tuning, and financial services experience including model risk and AI governance.