Data Scientist - Fraud Detection

About the role

Moniepoint, a fintech ranked among the fastest-growing in Africa, is looking for a Data Scientist to build the models, experiments and detection systems that protect its customers and merchants from fraud. The role sits at the intersection of machine learning, product and engineering, and involves prototyping and shipping ML models, designing experiments and uncovering new fraud signals in close partnership with engineers, product managers and analysts.

Key responsibilities

• Prototype, evaluate and help produce machine learning models for fraud detection, and own their ongoing monitoring and retraining cycles.
• Design and run experiments to measure the impact of fraud interventions, balancing customer experience against loss reduction.
• Size fraud typologies across product lines to inform prioritisation and investment decisions.
• Build and maintain anomaly detection systems to surface novel fraud vectors early.
• Work with fraud operations, engineers, product managers and data analysts to turn model outputs into real-world mitigations.

Required profile

• Degree in a quantitative field such as Statistics, Mathematics, Engineering or Computer Science.
• 3+ years of experience in data science, decision science or risk analytics within fraud, payments or financial crime.
• Hands-on experience building and deploying ML models in production.
• Fraud, risk or financial services experience is a strong plus.
• Comfortable working in fast-paced, cross-functional teams with high ownership expectations.

Required skills

• Solid grounding in data science fundamentals: experimentation, statistical inference, model evaluation and feature engineering.
• Proficiency in Python and SQL across the full model development lifecycle.
• Ability to communicate technical findings clearly to non-technical stakeholders.

What we offer

• An attractive salary, pension, health insurance and annual bonus.
• A learning and development environment with knowledge sharing, training and regular internal technical talks.