Senior AI Model Evaluation Director in London

Requirements:

• Bachelor's, Master’s, or PhD or equivalent experience in Computer Science, Machine Learning, Applied Mathematics, Statistics, Financial Engineering, or a related quantitative field.
• Significant professional experience (typically 7–12+ years) in AI/ML product development/management, model validation, quantitative research, risk modelling, or related areas.
• Demonstrated success working with technical teams or senior specialists in high-stakes modelling environments.
• Deep understanding of AI/ML systems—including LLMs, agentic architectures, RAG pipelines, credit or pricing models, or risk modelling techniques.
• Hands-on experience developing or validating models, performing statistical testing, and analysing model assumptions, limitations, and risks.
• Familiarity with model evaluation tooling, experimentation frameworks, and modern ML infrastructure.
• Excellent communication skills, with the ability to present complex findings clearly to both technical and non-technical audiences.
• (Desirable) Experience in B2B data or RegTech environments.
• (Desirable) Experience managing AI systems in production environments or high-scale data and ML platforms.
• (Desirable) Experience in working with teams in MLOps, DevOps, or large-scale compute environments (e.g., GPU clusters, cloud orchestration, Kubernetes).
• (Desirable) Experience with Generative AI evaluation, agent testing, or AI safety frameworks.
• (Desirable) Track record of partnering with regulatory bodies or leading audit-readiness efforts.

What the job involves:

We are seeking an experienced AI & Model Evaluation Manager to lead the evaluation, validation, and governance of advanced AI, machine learning, and statistical models across our Active Data Layer programme of development. This role blends technical depth, strategic leadership, and strong stakeholder management, ensuring that our models are accurate, reliable, safe, and aligned with regulatory and organisational standards.

You will oversee the end-to-end lifecycle of model evaluation - ranging from large language models (LLMs), agentic systems, and machine learning models used across Risk Intelligence to source, determine, match, and resolve model-driven data tasks within our Financial Crime and Screening domains.

You will act as the product owner of the AILab and associated analytics infrastructure supporting our ADL models, working in close partnership with technology leads on the development and implementation of these new capabilities. As a senior member of the Data & Product team, you will partner closely with engineering and architecture teams, including Data Science, AIOps, as well as the wider business functions such as Risk & Compliance, Legal, and Content Operations to ensure our data capabilities are accurate, fair, robust, auditable, and business-effective.

In addition, you will use your experience to shape best practices, drive innovation, and influence broader AI and model governance frameworks.

AI & ML Model Evaluation:

• Lead the design and execution of evaluation methodologies for LLMs, multimodal systems, AI agents, and traditional ML models.
• Oversee scenario-based testing, regression suites, multiturn agent simulations, and automated evaluation systems such as LLM-as-Judge and hybrid scoring approaches.
• Build, refine, and maintain frameworks that assess model quality, robustness, performance, safety, explainability, and reliability at scale.

Model Validation & Governance:

• Direct the independent review and validation of models across teams, ensuring compliance with internal LSEG governance standards and processes, and relevant regulatory expectations.
• Maintain a robust model inventory, validation documentation, and version controlled evidence supporting approval and audit requirements.
• Serve as a subject matter expert on model risk & decision methodologies (as applicable), AI evaluation patterns, and modelling frameworks.

Experimentation & Monitoring:

• Oversee the development and operation of online and offline experimentation platforms, including A/B testing, shadow deployments, canary releases, and continuous monitoring.
• Embed evaluation and experimentation into CI/CD pipelines, enabling automated quality gates and reliable release processes for model-driven products.
• Implement observability practices that track model drift, degradation, safety issues, and agent behaviour over time.

Leadership & Strategy:

• Work collaboratively within a cross-functional high-performing team, fostering innovation, technical excellence, and a collaborative culture.
• Define and execute strategic direction for AI evaluation, model risk management, and model governance frameworks.
• Partner with product, engineering, research, risk, compliance, and senior leadership across the organisation to influence AI development practices and decision-making.
• Represent the function in internal and external audits, regulatory engagements, and cross-functional governance forums.

Stakeholder Engagement:

• Act as a trusted advisor to model owners, developers, and business leaders—translating complex technical findings into actionable insights.
• Support change management across the organisation to drive consistency in evaluation standards, documentation quality, and responsible AI adoption.

What Success Looks Like:

• Models across the organisation consistently meet high standards of quality, safety, performance, and compliance.
• The business benefits from reliable model-driven decisions, underpinned by transparent, well-governed evaluation practices.
• The team operates with excellence, delivering high-impact insights, innovative evaluation techniques, and robust model validation outcomes.
• Stakeholders across engineering, risk, and product view the function as a strategic partner and trusted authority.