Senior Data Scientist - Credit Risk & BI

About the role

Pezesha, an embedded finance company providing credit solutions to MSMEs across Africa, is hiring a Senior Data Scientist in Nairobi (hybrid/remote options available). The role reports to the Data Science Manager and leads the development of credit risk models, strategic insights and business intelligence tools that drive repayment rates, operational efficiency and portfolio growth.

Key responsibilities

• Develop, test and iterate credit risk models quarterly, keeping repayment rates above 95% and default rates below 3% across all portfolios.
• Provide data-driven insights and reports to portfolio leads, senior management and external stakeholders, turning findings into actionable strategies.
• Lead data extraction, cleaning, transformation and mining, building datasets for descriptive, predictive and spatial models.
• Design and run experiments across business units to optimise products, processes and risk strategies.
• Develop and enhance BI tools, delivering at least two major enhancements or new features per quarter.
• Build and maintain real-time dashboards with 99% uptime and accuracy, deliver monthly finance reports on time with greater than 99% accuracy, and ensure accurate lender reporting and portfolio impact analysis.
• Maintain clear documentation of models, experiments and workflows, mentor junior data scientists and analysts, and set priorities and manage workloads for the data science team.
• Execute ad hoc tasks assigned by the CEO, COO, Data Science Manager or CDO.

Required profile

• Master's degree or higher in Data Science, Statistics, Computer Science, Economics or a related field.
• 6+ years of proven experience in data science, machine learning and credit risk modelling, preferably in fintech or financial services.
• Strong understanding of credit risk management, repayment behaviour and portfolio analytics.
• Leadership, mentoring and cross-functional collaboration skills, with strong communication abilities.

Required skills

• Python, R, SQL and big data tools.
• Machine learning frameworks such as TensorFlow, Scikit-learn and XGBoost.
• Data visualisation tools such as PowerBI, ggplot or plotly.

Performance targets include LMS data validation accuracy of at least 98%, sprint completion of 90%+ on time, SLA adherence of 95%+, quarterly knowledge transfer sessions, and at least two BI enhancements per quarter. To apply, send your CV and GitHub link or work sample by email to the company with the reference REF 26.