AI & General Model Validation (AIGMV) provides independent technical challenge and review of the Groups most important ML and AI models, across business and support units. As a Senior Data Scientist within the team, you will independently review, challenge and validate AI use cases across the Group. You will assess whether AI systems are fit for purpose and operate as intended, and whether the risks they introduce are understood and appropriately controlled. 

You will design and carry out validation that is proportionate to the risk, covering data, methodology, assumptions and outcomes, and explain your findings clearly to technical and non-technical stakeholders. Model approvals, usage decisions and risk acceptance will draw directly on your analysis and judgement. 

You will work closely with model developers, product owners and business stakeholders, while remaining independent of them, and provide constructive challenge that supports safe and sustainable AI adoption. You will also help develop the teams AI validation practices, tooling and standards, sharing knowledge and building capability across the team. 

• A courageous mindset and a proven record of self-directed learning, with the curiosity and discipline to stay ahead of fast-changing AI techniques and risks 

• Close attention to detail and a knack for spotting problems 

• Working knowledge of GenAI and agentic architecture, including orchestrators, sub-agents, RAG pipelines, tool use, guardrails, multi-agent routing, memory and persona handling, and LLM-as-a-Judge 

• Hands-on experience designing, building and evaluating AI and Generative AI solutions, including synthetic data generation. Experience in fine-tuning transformer models is also a plus. 

• Experience with current GenAI frameworks and tools like LangChain, OpenAI Agents SDK, LangGraph, Pydantic, Langfuse/HoneyHive or similar platforms. 

• Advanced Python, plus strong experience with GitHub, version control and CI/CD pipelines. SQL, R or TypeScript is a plus. 

• Comfort reading and navigating unfamiliar codebases, reproducing evaluation pipelines, writing test harnesses and inspecting trace data 

• Hands-on experience with AI coding agents, including Claude Code and Codex, and the ability to build and maintain skills, sub-agents, hooks and plugins that automate workflows. 

• Working knowledge of traditional machine learning models, including regression, random forests, XGBoost, and related evaluation methods.  

• Strong communication skills, with the ability to explain complex technical concepts to mixed audiences.