Senior Data Scientist - Stress Testing Models

About the role

The Model Development Chapter Area sits within Risk Management and develops credit risk models and advanced analytics across the group. As a Senior Data Scientist - Stress Testing Models, you will lead the development and enhancement of credit risk stress-testing models used across the group, covering retail and non-retail portfolios. You will work with colleagues in model development, independent model validation, business unit risk, Finance and Enterprise Services.

Key responsibilities

• Build statistical models and run advanced analysis using predictive modelling, non-linear regression, time-series analysis, panel-data methods and macroeconomic modelling.
• Translate macroeconomic and portfolio scenarios into credit risk impacts, including changes in Probability of Default (PD), Loss Given Default (LGD), Exposure at Default (EAD) and expected credit loss.
• Analyse model behaviour, sensitivities, limitations and key risk drivers to interpret stress-testing results.
• Work with large, complex and sometimes imperfect datasets, applying sound judgement to data challenges.
• Use R, Python and SQL, alongside Teradata and AWS, to prepare, transform and analyse data for modelling.
• Articulate modelling methodologies, assumptions, limitations and conclusions so they can be understood, challenged and validated.
• Partner with model users, validation, risk teams and Finance throughout the modelling lifecycle.
• Provide technical guidance and constructive review to other analysts, and explore responsible uses of AI to improve analysis and productivity.

Required profile

• Strong experience in quantitative statistical modelling, data science, econometrics or credit risk analytics, able to solve complex modelling problems independently.
• Experience in credit risk modelling (PD, LGD, EAD, stress testing, IFRS 9 or IRB) highly regarded.
• Experience leading analytical work, influencing stakeholders and guiding less experienced colleagues.
• Tertiary qualification in a quantitative discipline such as Mathematics, Statistics, Econometrics, Engineering or Data Science.

Required skills

• Statistical, econometric or predictive modelling techniques, including logistic and non-linear regression, time-series analysis, panel modelling, macroeconomic modelling and machine learning.
• Hands-on R and/or Python, SQL and experience with large datasets.
• Strong written and verbal communication, explaining methodology and conclusions to technical and non-technical audiences.