SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Allica Bank is the UK's fastest-growing fintech firm, focused on serving established SMEs—a major underserved segment in commercial lending. The Modelling team within Credit Portfolio Management develops measurement and decisioning capabilities that power Allica's lending products, including IFRS9 models (PD, LGD, SICR), statistical and machine learning models, credit policy rules, and GenAI-driven lending solutions.
In this role, you will take a hands-on approach to designing, building, implementing, and optimizing the statistical and machine learning models that drive automated lending decisions. You'll work across the full model lifecycle: data preparation, feature engineering, model fitting, validation, implementation, and ongoing performance monitoring. Your responsibilities include developing and deploying credit risk models in Python, performing ad-hoc analysis to optimize credit policy rules, producing regular scorecard and decisioning monitoring reports (PSI, Gini, KS, population stability, calibration), and automating monitoring packs and controls.
You'll lead root-cause investigations into model or decisioning underperformance by tracing issues through decision logs, data lineage, and production pipelines. You'll recalibrate or redevelop models as portfolio experience matures, comparing predicted versus actual outcomes and adjusting thresholds where drift or mis-calibration is detected. Building robust, reproducible modelling processes and driving improvements to data, tooling, and infrastructure are core to the role. You'll document models, assumptions, and changes to support governance and independent validation, and collaborate closely with Credit Risk, Underwriting, Product, Data, and Engineering teams to translate business requirements into practical decisioning solutions while maintaining strong internal controls.
Required: at least 4 years of hands-on risk modelling experience, proficiency in statistical techniques (logistic/linear regression) and machine learning, comfort with large datasets and SQL/Python/BI tools, proven Python coding ability, understanding of credit and commercial lending concepts, and strong technical and business communication skills. Experience implementing models into software applications is highly desirable. You are self-motivated, adaptable, and collaborative.
