The posting
Role Overview
We are seeking experienced Data Scientists to support the maintenance, enhancement and expansion of fraud-risk analytics models designed to detect potential misuse of public grants.
The successful candidates will be responsible for reviewing and improving existing fraud detection methodologies, recalibrating model thresholds, developing performance monitoring dashboards and assessing opportunities to extend fraud-risk coverage to additional programmes.
This is a hands-on role involving fraud analytics, predictive modelling, model validation, performance monitoring and stakeholder collaboration to deliver reliable, effective and sustainable fraud detection solutions.
Key Responsibilities
- Review and recalibrate existing fraud detection models, methodologies and risk thresholds to improve detection accuracy and effectiveness.
- Conduct model back-testing, impact assessments, gap analyses and validation to evaluate model performance and identify areas for improvement.
- Translate business, policy and operational requirements into enhancements to fraud-risk scenarios, indicators, features, thresholds and analytical logic.
- Design, develop and enhance technical dashboards to monitor model performance, including precision, recall, data drift and concept drift.
- Implement monitoring metrics, alert mechanisms and visualisations to identify model performance deterioration and emerging fraud risks.
- Develop model evaluation approaches that account for delayed, incomplete or evolving fraud investigation outcomes.
- Assess new programmes and schemes, including eligibility criteria, claims processes and potential fraud vulnerabilities.
- Analyse fraud patterns and risk indicators to recommend suitable detection methodologies and model enhancements.
- Evaluate opportunities to extend existing fraud detection models to additional programmes and recommend implementation approaches.
- Collaborate with business stakeholders, programme owners, data engineers and management to communicate analytical findings, risks and recommendations.
- Maintain clear documentation of model methodologies, threshold adjustments, validation results and technical decisions.
- Develop reproducible analytical code, technical documentation and knowledge-sharing materials to support ongoing model maintenance.
Requirements
Education & Experience
- Bachelor's degree in Data Science, Statistics, Mathematics, Computer Science, Engineering, Economics or a related quantitative discipline.
- 3–5 years of hands-on experience in data science, advanced analytics, model validation, fraud/risk analytics or programme analytics.
Technical Requirements (Must-Have)
- Strong proficiency in Python and SQL for data extraction, transformation, exploratory data analysis and reproducible analytical workflows.
- Hands-on experience developing, validating or enhancing predictive models, risk-scoring models, fraud detection models or anomaly detection models.
- Experience in feature engineering, model back-testing, threshold calibration and performance evaluation.
- Strong understanding of model evaluation metrics, including precision, recall and associated performance trade-offs.
- Ability to translate business, policy and operational requirements into analytical specifications, model logic, risk indicators and measurable hypotheses.
- Experience working with complex, large-scale datasets, including data quality assessment, data lineage, anomaly investigation and documentation of analytical assumptions.
Preferred Qualifications (Good-to-Have)
- Experience handling imbalanced datasets and delayed, incomplete or evolving outcome labels in fraud or risk modelling.
- Experience implementing model monitoring solutions, including data drift, concept drift and alert thresholds.
- Experience developing analytical dashboards using Tableau or equivalent visualisation platforms.
- Familiarity with CI/CD, MLOps practices and operational deployment of analytical models.
- Experience with Apache Spark, PySpark, AWS Redshift and AWS Glue.
- Experience with network or graph analytics to identify related entities, suspicious relationships or coordinated fraud patterns.
- Experience analysing workforce, training, grants, claims or regulatory datasets.
- Prior experience working on Singapore public-sector projects or familiarity with government data classification requirements.
Soft Skills
- Strong analytical thinking, structured problem-solving and attention to detail.
- Ability to translate ambiguous requirements into clear analytical methodologies, assessment plans and implementation approaches.
- Excellent written and verbal communication skills, with the ability to explain technical findings and recommendations to non-technical stakeholders.
- Ability to work independently while collaborating effectively with data scientists, engineers, platform teams and business stakeholders.
- Strong organisational skills with the ability to manage delivery risks, dependencies and competing priorities.
- Proactive mindset with a focus on analytical accuracy, continuous improvement and reliable delivery.
