Contract Duration: 3 Months
Engagement: Senior-Level Consulting / Contract
Workstream: Module B – Behavioural Analytics & Machine Learning
About the Role
We are seeking an experienced Data Scientist to support the development and enhancement of fraud detection and behavioural analytics capabilities.
The successful candidate will focus on behavioural and historical interaction feature engineering, model selection, anomaly detection, model drift and performance benchmarking.
The ideal candidate will have strong experience working with transactional data and applying unsupervised machine learning techniques to identify unusual patterns and potential fraudulent behaviour.
Key Responsibilities
- Support the delivery of Module B – Behavioural Analytics & Machine Learning.
- Develop behavioural and historical interaction features from transactional data.
- Conduct feature engineering and feature evaluation.
- Review and enhance the existing MO/scheme library.
- Assess and recommend appropriate machine learning models.
- Apply unsupervised anomaly detection techniques to transactional datasets.
- Analyse historical behavioural patterns and identify unusual or suspicious activity.
- Support model experimentation, validation and selection.
- Monitor model performance and identify potential model drift.
- Develop approaches for drift detection and model control.
- Conduct before-and-after performance benchmarking.
- Analyse model outputs and provide actionable recommendations.
- Collaborate with the Senior Graph Data Scientist / ML Lead.
- Work closely with the MLOps Engineer to support productionisation.
- Document methodologies, features, model decisions and performance results.
Required Experience & Skills
- Proven experience as a Data Scientist, preferably in fraud analytics or financial services.
- Strong Python and machine learning experience.
- Strong experience in feature engineering.
- Experience working with transactional and behavioural datasets.
- Hands-on experience with unsupervised anomaly detection.
- Experience with model selection, validation and performance evaluation.
- Understanding of model monitoring and model drift.
- Experience conducting model benchmarking and performance analysis.
- Strong statistical and analytical skills.
- Ability to translate complex datasets into meaningful behavioural insights.
- Strong communication and documentation skills.
Advantageous
- Experience in banking, payments, financial services or fintech.
- Experience in fraud detection and transaction monitoring.
- Experience with graph analytics or graph-derived features.
- Experience with real-time or near-real-time analytics.
- Experience with scheme/MO libraries or fraud rule engines.
- Experience working within regulated environments.
Engagement
This is a 3-month senior-level engagement supporting the broader fraud detection and machine learning workstream.
Ideal candidate: A hands-on Data Scientist with strong transactional analytics and anomaly detection experience who can improve feature engineering, model performance and fraud detection outcomes.