The posting
Data Scientist – Fraud Analytics / Risk Modelling
Role Overview
We are seeking an experienced Mid to Senior Level Data Scientist to enhance and maintain fraud detection and risk analytics models. The role involves model validation, threshold optimisation, performance monitoring and developing analytical solutions to identify potential fraud risks across public training grant schemes.
Key Responsibilities
- Review, validate and optimise fraud detection models, including methodology assessments, backtesting, threshold calibration and performance evaluation.
- Develop and enhance fraud-risk indicators, scoring models, analytical rules and features based on business and operational requirements.
- Build and maintain technical dashboards to monitor model performance, including precision, recall, data drift, concept drift and alert thresholds.
- Analyse new grant schemes, eligibility criteria and claims processes to identify potential fraud risks and recommend appropriate detection approaches.
- Perform exploratory data analysis, data validation, anomaly investigation and impact assessments using large-scale datasets.
- Collaborate with business stakeholders, data engineers and management to translate requirements into analytical solutions.
- Maintain reproducible analytical workflows, model documentation, validation reports and knowledge transfer materials.
Requirements
- Degree in Data Science, Statistics, Mathematics, Computer Science, Engineering, Economics or a related quantitative discipline.
- Min 4 years of experience in data science, advanced analytics, fraud/risk modelling or model validation.
- Strong hands-on proficiency in Python and SQL for data analysis, transformation and model development.
- Experience developing, validating or enhancing predictive models, risk-scoring models, fraud detection or anomaly detection solutions.
- Strong understanding of feature engineering, backtesting, threshold calibration and model performance metrics, including precision and recall.
- Experience working with large, complex datasets, including data quality assessment, anomaly detection and data lineage.
- Ability to translate business requirements into analytical specifications, model logic and measurable risk indicators.
- Strong analytical, problem-solving, documentation and stakeholder communication skills. Exposure to workforce, training grants, claims, regulatory datasets or Singapore public-sector projects.
Good to Have
- Experience handling imbalanced datasets, delayed outcome labels and fraud/risk model monitoring.
- Knowledge of data drift, concept drift and production model performance monitoring.
- Experience with Tableau or equivalent dashboarding tools.
- Familiarity with CI/CD, MLOps and model deployment.
- Experience with Spark, PySpark, AWS Redshift and AWS Glue.
- Knowledge of graph/network analytics for identifying suspicious relationships or coordinated fraud patterns.



