About the position
Senior Data Scientist – Graph Analytics, AI and Financial Crime
Role purpose
To design, develop, deploy and optimise advanced analytics, machine-learning and graph-based solutions that generate actionable insight from complex, interconnected enterprise data.
The Senior Data Scientist will support high-impact banking use cases across fraud detection, financial crime, customer intelligence and AI/GenAI enablement. The role will be accountable for end-to-end model development, graph analytics, feature engineering, performance optimisation and production deployment within the bank's enterprise data and cloud environments.
Key responsibilities
Advanced analytics and machine learning
- Design, build and deploy predictive, classification, anomaly-detection and clustering models for complex business and risk use cases.
- Apply machine-learning techniques to identify fraud patterns, suspicious behaviour, financial-crime risk indicators and customer relationship insights.
- Develop robust model features from structured, semi-structured and interconnected data sources.
- Perform exploratory data analysis, data profiling, hypothesis testing and model validation.
- Evaluate, tune and optimise models for accuracy, scalability, interpretability and operational performance.
- Establish appropriate model-monitoring approaches, including drift detection, performance tracking and retraining considerations.
Graph analytics and knowledge graphs
- Design and develop graph-based data models representing relationships between customers, accounts, transactions, devices, merchants, organisations and related entities.
- Apply graph analytics techniques to identify hidden relationships, communities, network anomalies, suspicious transaction patterns and connected-risk indicators.
- Use graph algorithms such as centrality, similarity, path analysis, community detection, link prediction and entity resolution where relevant.
- Create graph-derived features for use in downstream machine-learning and risk models.
- Build and maintain knowledge-graph capabilities that enable AI and GenAI use cases, including improved retrieval, context enrichment, entity relationships and semantic understanding.
- Collaborate with data engineering and architecture teams to ensure graph data is scalable, governed and production-ready.
AI and GenAI enablement
- Support the development of AI and GenAI solutions through graph-enhanced data structures, semantic models and knowledge graphs.
- Contribute to retrieval-augmented generation, context enrichment and relationship-aware AI use cases where required.
- Partner with AI engineers, data engineers and solution architects to operationalise AI capabilities securely within the enterprise environment.
- Ensure AI and data-science solutions are aligned with responsible AI, data governance, privacy and model-risk requirements.
Data, platform and deployment accountability
- Work with data engineers to source, prepare and integrate data from enterprise platforms, transactional systems, APIs and external data sources.
- Develop reusable Python code, feature pipelines, model components and analytical assets.
- Package and deploy models using cloud-native, containerised or MLOps-aligned delivery practices.
- Contribute to model operationalisation, CI/CD pipelines, version control, documentation and production support.
- Optimise solutions for performance, scalability, reliability and maintainability.
- Ensure data quality, lineage, reproducibility and auditability across analytical and model-development processes.
Stakeholder engagement and delivery
- Translate business, fraud, financial-crime and customer-intelligence requirements into analytical problem statements and measurable data-science solutions.
- Communicate model outcomes, limitations, assumptions and recommendations clearly to technical and non-technical stakeholders.
- Partner with risk, compliance, fraud, AML, customer, data and technology teams to identify high-value use cases.
- Mentor junior data scientists and contribute to data-science standards, reusable assets and technical best practice.
Required experience
- 7+ years’ experience in data science, advanced analytics, machine learning or related quantitative roles.
- Proven experience developing and deploying machine-learning models in production or enterprise environments.
- Strong Python development capability for data science, modelling, feature engineering and automation.
- Experience with machine-learning libraries such as scikit-learn, XGBoost, LightGBM, PyTorch or TensorFlow.
- Demonstrable experience with graph analytics, graph data modelling or graph databases.
- Experience applying graph techniques to real-world use cases such as fraud, financial crime, entity resolution, network analysis, customer intelligence or recommendation engines.
- Experience building graph-derived features for predictive or risk models.
- Experience working with large, complex and interconnected datasets.
- Strong SQL and data-wrangling capability.
- Experience working in cloud-native data platforms and modern data-engineering ecosystems.
- Exposure to MLOps, model deployment, CI/CD, version control and model monitoring.
- Ability to work independently across the full data-science lifecycle, from problem definition to deployment.
Preferred technical skills
- Graph databases and tools such as Neo4j, TigerGraph, Amazon Neptune, Azure Cosmos DB Gremlin API, JanusGraph or similar.
- Graph analytics libraries such as NetworkX, Neo4j Graph Data Science, PyTorch Geometric, DGL or equivalent.
- Knowledge-graph technologies, including RDF, SPARQL, ontologies, semantic modelling or entity-resolution frameworks.
- Experience with cloud platforms such as Azure, AWS or Google Cloud.
- Exposure to Databricks, Spark, Snowflake, Kubernetes, Docker, MLflow, Azure Machine Learning, SageMaker or similar platforms.
- Experience with fraud, AML, transaction monitoring, sanctions, KYC, credit risk or financial-crime analytics.
- Knowledge of GenAI, retrieval-augmented generation and graph-enhanced AI architectures.
- Experience working with streaming or near-real-time data pipelines.
Qualifications
- Degree in Data Science, Computer Science, Statistics, Mathematics, Engineering, Actuarial Science, Quantitative Finance or a related field.
- Postgraduate qualification in a relevant quantitative discipline would be advantageous.
- Relevant cloud, machine-learning, graph database or data-science certifications would be beneficial.
Desired Skills:
- Senior Data Scientist
- Lead Data Scientist
- Principal Data Scientist
