Data Scientist - Fraud Analytics

Client of Capitex

Role Overview

We are seeking a highly analytical and experienced Data Scientist Fraud Analytics to join our team on a 12-month contract in Kuwait. The successful candidate will be responsible for developing, enhancing, and optimising fraud detection models and analytics strategies across digital and payment channels.

This role will focus on leveraging advanced analytics, machine learning, and statistical techniques to strengthen fraud detection capabilities, reduce losses, and improve customer experience by minimising false positives.

Key Responsibilities

• Develop, validate, and deploy fraud detection models across digital banking and payment systems.
• Perform advanced data analysis to identify fraud patterns, emerging threats, and behavioural anomalies.
• Optimise existing fraud detection strategies through model tuning and performance monitoring.
• Conduct feature engineering and model performance evaluation using appropriate metrics (e.g., precision, recall, AUC, false positive rate).
• Collaborate with fraud risk, rule writing, and technology teams to translate analytical insights into actionable controls.
• Support model governance processes including documentation, validation, and regulatory compliance requirements.
• Analyse large datasets to uncover trends and recommend improvements to fraud prevention strategies.
• Assist with model implementation, testing (UAT), and post-deployment performance monitoring.
• Stay current with emerging fraud typologies and advancements in machine learning techniques.

Required Skills & Experience

• Proven experience in fraud analytics and model development within banking, fintech, or financial services.
• Strong knowledge of machine learning techniques (e.g., logistic regression, decision trees, random forests, gradient boosting, neural networks).
• Hands-on experience with Python, R, or similar analytical programming languages.
• Strong SQL skills and experience working with large transactional datasets.
• Understanding of fraud typologies including account takeover, card-not-present fraud, mule accounts, and social engineering.
• Experience with model performance monitoring and optimisation.
• Strong analytical thinking and problem-solving skills.