Data Scientist
Summary
Data Scientist for a financial services and insurance group in Johannesburg, turning business questions into models across finance and actuarial data — forecasting, cost analytics, anomaly detection and reporting intelligence. Core stack: Python/R, SQL, scikit-learn/XGBoost/TensorFlow, Azure ML/Databricks/SageMaker, Informatica, Power BI/Tableau.
Our client, a leading financial services and insurance group, is looking for a Data Scientist to apply advanced analytics, statistical modelling and machine learning to finance and actuarial data. Reporting to the Head: Centre IT, you will turn business questions into analytical problems, build and productionise models, and translate results into insight that shapes decision-making across the finance operating model, from forecasting and cost analytics to anomaly detection, automation and reporting intelligence.
Key focus areas: problem framing and exploratory analysis, model development and MLOps deployment, analytical data products, visualisation and stakeholder storytelling, and model risk, ethics and governance.
Requirements
- 6+ years in data science, advanced analytics or quantitative modelling, with 3+ years in insurance or financial services
- Advanced Python and/or R, plus strong SQL for large-scale data manipulation
- Practical ML experience with scikit-learn, XGBoost, TensorFlow or PyTorch
- Solid statistics: regression and GLMs, time-series forecasting, hypothesis testing, experimental design
- Model deployment to production on Azure ML, Databricks, AWS SageMaker or equivalent, with MLOps tooling
- Strong grasp of finance data flows, transformation and cleansing; Informatica or comparable ETL
- Familiarity with Data Mesh, MDM and finance data lakes/warehouses
- Visualisation and storytelling in Power BI, Tableau or equivalent
- Understanding of finance and actuarial data, accounting principles and reporting standards
- Git, Jira; Agile and Waterfall
