Data Scientist - Digital Payments (Ghana Pilot)

About the role

A 6-month on-site contract Data Scientist position in Accra, Ghana, supporting a Ghana pilot project. The Data Scientist turns complex banking and payments data into practical insights, predictive models and customer segments, working closely with technical and business stakeholders to prepare data, validate analytical approaches and turn model outputs into implementation recommendations. Travel across West and Central Africa is required.

Key responsibilities

• Acquire, profile, clean and transform large transactional, customer and payments datasets.
• Develop and validate predictive, classification or propensity models for business and implementation use cases.
• Build customer segmentation approaches using transactional and behavioural indicators.
• Perform feature engineering, exploratory analysis and statistical validation, documenting assumptions and limitations.
• Create reproducible analytical workflows and share findings through visualisations and stakeholder-ready summaries.
• Partner with engineering and integration teams on data inputs, quality checks and deployment requirements.
• Apply appropriate controls when handling sensitive banking and payments data.

Required profile

• Bachelor's degree in Finance, Economics, Business Administration, Computer Science, Information Technology or a related field (Master's preferred).
• Previous experience in banking, digital payment and card solutions or the FinTech industry.
• Regional experience, including in a fast-paced matrix organisation.
• Excellent English (French is an advantage); willingness to travel across West and Central Africa may be relevant for later phases.

Required skills

• Python, SQL, pandas, scikit-learn, Jupyter, Power BI or Tableau, Git and cloud data platforms.
• Building and validating predictive, classification or segmentation models.
• Understanding of model evaluation, bias, overfitting, data quality and reproducibility.
• Ability to explain analytical methods and results to technical and non-technical audiences.

Preferred: payment transactions, card portfolios or customer lifecycle analytics; model deployment, MLOps or dashboarding; campaign measurement, experimentation or customer targeting. Deliverables include a discovery and implementation plan, high-quality outputs with testing and documentation, timely risk escalation, and reusable implementation assets and knowledge transfer.