Location
London

Hours
Full Time

Salary
Negotiable

About the Role
At Moody's, we unite the brightest minds to turn today’s risks into tomorrow’s opportunities. We strive to create an inclusive environment where everyone feels welcome to be who they are—with the freedom to exchange ideas, think innovatively, and listen to each other and customers in meaningful ways. Moody’s is transforming how the world sees risk. As a global leader in ratings and integrated risk assessment, we’re advancing AI to move from insight to action—enabling intelligence that not only understands complexity but responds to it. We decode risk to unlock opportunity, helping our clients navigate uncertainty with clarity, speed, and confidence.
If you are excited about this opportunity but do not meet every single requirement, please apply! You still may be a great fit for this role or other open roles. We are seeking candidates who model our values: invest in every relationship, lead with curiosity, champion diverse perspectives, turn inputs into actions, and uphold trust through integrity.

The role carries out the independent evaluation and challenge of models, scorecards and agents used in the context of credit rating activities across asset classes, as well as AI model risk management. The successful candidate will bring strong, practical expertise in Generative and agentic AI and play an active role in advancing the organization’s approach to evaluating and governing these systems.

Key responsibilities include:
- Carrying out independent evaluation and challenge of AI-enabled tools, agents and quantitative models, including detailed assessment of inputs, assumptions, conceptual soundness, performance and limitations, building credit model-validation rigor with the support of the wider team
- Performing AI model risk management activities, including identifying, assessing, and mitigating risks associated with Generative AI and agentic AI
- Designing and delivering complex validation analyses, including model replication, challenger development, sensitivity testing, benchmarking, and ad-hoc quantitative investigations
- Delivering validation outputs, from test plan design through clear, concise validation reporting and effective challenge of model developers
- Maintaining clear separation between model development and independent review, preserving the independence of Model Risk Group’s conclusions
- Building AI model risk capability within the team, sharing Generative AI and agentic AI expertise, advancing evaluation capabilities, and developing subject-matter expertise across analytical and methodology teams
- Driving rigorous execution standards and culture, promoting effective challenge, continuous improvement, and practical application of model risk frameworks and policies

About the team:
The Model Risk Group Quantitative Review team independently reviews and validates quantitative models and scorecards that support credit ratings. It assesses whether these tools are conceptually sound, appropriately calibrated, fit for purpose, and operate in line with approved methodologies and governance standards.

Moody’s is an equal opportunity employer. All qualified applicants will receive consideration for employment without regard to race, color, religion, sex, national origin, disability, protected veteran status, sexual orientation, gender expression, gender identity or any other characteristic protected by law.

Experience
- Strong understanding of AI model risk management, including risks and controls specific to Generative AI and agentic AI, and their implications for model validation and governance
- Practitioner-level agentic AI skills with a demonstrated track record of building, deploying and evaluating AI agents and multi-agent systems, including tool use, orchestration and state management using modern agent frameworks
- Strong, practical knowledge of agent evaluation techniques such as task-success and trajectory testing, grounding and retrieval quality assessment, hallucination and drift detection, robustness testing, benchmarking, bias and fairness assessment, alignment and misalignment analysis, adversarial testing, and red-teaming
- Strong LLM/Generative AI engineering fluency including prompting strategies, retrieval-augmented generation, context and memory design, function/tool calling, guardrails, observability, and failure-mode analysis
- Proven track record of taking advanced AI systems from prototype to production with attention to reliability, safety, and responsible AI controls
- Solid understanding of quantitative finance, modeling and model validation, together with financial products and markets is valued but not essential for candidates with exceptional AI and agentic depth
- Proficiency in programming languages such as R, Python, MATLAB, and SQL, with ability to work within an established codebase; knowledge of C++ preferred
- Experience in model validation, model risk management or independent review is an advantage; domain expertise in credit model risk can be developed through collaboration with experienced validators
- Relevant professional or research experience spanning AI, machine learning, quantitative analytics, software engineering, model risk management or related disciplines; practical skill with advanced AI and agentic systems prioritized over years of experience

About you
- Drive and ownership: bring energy and initiative, take ownership of your work, build trusted relationships, and role model a growth mindset and effective challenge across the team
- Highly organized, efficient, and detail-oriented, able to prioritize competing demands, work to tight deadlines, and deliver accurate, high-quality outputs both independently and collaboratively
- Strong communicator, able to articulate complex ideas fluently and clearly for both technical and non-technical audiences, with strong written and spoken English

Qualifications
- Strong academic background in a technical or quantitative field such as computer science, artificial intelligence, machine learning, mathematics, physics or engineering
- Preference for candidates holding a post-graduate degree
- Demonstrated, applied agentic AI capability is weighted alongside formal credentials