Data Scientist – Decision Science & Modelling

About this role

We are seeking a Data Scientist to design and deliver analytical models and decision-support systems that improve understanding, prediction and decision-making across the business. The role focuses on building practical models of complex real-world systems, working with imperfect data, uncertainty and competing objectives to generate commercially valuable outcomes. Responsible for developing deployable analytical solutions in partnership with Engineering teams, while not owning production infrastructure or application development.

Key Responsibilities: Model Development Design, develop and maintain statistical, probabilistic and simulation-based models. Translate complex business questions into tractable modelling problems. Select appropriate modelling approaches based on the characteristics of the problem rather than methodological preference. Develop prototypes and working solutions iteratively, refining approaches as new information becomes available.

Build analytical assets that can be reused as products, decision-support tools or operational capabilities. Design and analyse experiments to evaluate interventions, operational changes and model effectiveness. Inference & Uncertainty Work effectively with incomplete, imperfect and evolving datasets. Develop approaches for estimating missing information and combining evidence from multiple sources. Quantify uncertainty and communicate appropriate confidence in model outputs.

Test assumptions and identify limitations within modelling approaches. Optimisation & Decision Support Develop frameworks that improve operational and commercial decision-making. Evaluate alternative actions, trade-offs and potential outcomes. Support automation of appropriate decision processes through analytical models. Design experiments and simulations that inform strategic and operational choices. Validation & Quality Validate models using appropriate testing, back-testing and comparison techniques.

Assess sensitivity to assumptions and changing conditions. Monitor model performance over time and identify concept drift or degradation. Maintain high standards of analytical rigour and reproducibility. Collaboration & Communication Partner closely with Business Analysts, Engineering, Product and operational teams. Explain modelling approaches, assumptions and results clearly to non-technical audiences. Document methodologies, limitations and recommendations in a practical and accessible way.