Data Scientist

The Opportunity

We are looking for two experienced Data Scientists to join a high-impact financial crime programme, working alongside a specialist delivery team to design, build and productionise AI and machine learning solutions at scale.

This is not a research role. You will be building and deploying production-grade models that directly support financial crime detection and supervision. The work is technically demanding, commercially grounded, and consequential.

A background in financial crime, RegTech or data-led supervision is essential. You will be operating in a regulated environment where model quality, explainability and auditability are not optional extras.

• Design and develop AI and ML solutions for financial crime detection and risk prioritisation, including classification, outlier detection and ranking models.
• Build and deploy production-level solutions in collaboration with a team of data scientists, ensuring code is clean, reproducible and maintainable.
• Develop and deploy containerised ML services, working within established pipeline and infrastructure frameworks.
• Conduct exploratory data analysis to identify early signals, risk clusters and emerging trends across financial data.
• Apply time-series analysis to assess risk patterns and changes in client behaviour over time.
• Troubleshoot, debug and optimise existing models and code under production conditions.
• Work across teams to understand business problems and translate them into effective, scalable data science solutions.
• Produce clear documentation of models, methodologies and outputs to support audit, governance and regulatory requirements.

• A demonstrable background in financial crime, RegTech or data-led supervision. This is a core requirement, not a bonus. You understand the regulatory context in which these models operate and the standards they must meet.
• 3 to 5 years of hands-on experience in data science, with in-depth knowledge of your specialist field.
• Strong Python skills across pandas, NumPy and scikit-learn for data wrangling, feature engineering and modelling.
• Solid SQL capability for querying structured data sources.
• Proven experience developing and validating classification, unsupervised learning and ranking models.
• Familiarity with containerised ML deployment, including tools such as Podman, SageMaker or DSW pipelines.
• Proficient use of Git for version control and collaborative, reproducible workflows.
• Experience with time-series analysis to assess risk trends across financial data.
• Strong exploratory data analysis skills with the ability to identify early signals and risk clusters from complex datasets.

• Experience with rank aggregation and ensemble techniques, including methods such as Robust Rank Fusion.
• Familiarity with model explainability tools such as SHAP or LIME to support interpretability in regulated environments.
• Experience with model monitoring and drift detection in production settings.
• Experience with record linkage or network analytics tasks.
• Knowledge of graph query languages such as Gremlin or Cypher, graph database platforms such as Neptune or Neo4j, or graph visualisation tooling.