Started in 2018 with one specific problem

Financial analysts in KwaZulu-Natal kept asking the same question: how do I actually use regression models on my loan data? They'd taken Python courses, watched YouTube tutorials, read documentation. But when they opened their spreadsheets full of customer transactions and credit scores, they didn't know where to start.

So we built webinars that begin with their datasets. Real CSV files from banks, insurance companies, investment firms. We show how to clean messy data, handle missing values, split training sets, validate models, and interpret results that matter to their managers. Not academic exercises—actual workflows that financial professionals use every day.

The approach worked because it skipped the usual problem: spending weeks on fundamentals before touching real data. Our participants start with their actual analysis challenges on day one. They learn scikit-learn functions by applying them to fraud detection. They understand cross-validation by testing credit scoring models. Theory comes when they need it to solve something specific.

What you get from our sessions

Each webinar focuses on one supervised learning technique applied to one financial problem. Logistic regression for default prediction. Random forests for customer segmentation. Gradient boosting for portfolio risk assessment. You work through examples with datasets similar to what you see at work.

• Live coding sessions where you follow along in Jupyter notebooks with financial datasets
• Explanations of when each algorithm works and when it fails for different analysis types
• Code templates you can adapt to your specific data structures and business requirements
• Direct answers to technical questions about your actual projects during Q&A segments
• Access to recorded sessions and example notebooks for reference when building your models

Who benefits most from this approach

Our webinars work best for financial analysts, risk managers, and portfolio specialists who already understand their data and business problems. You know what insights your stakeholders need—you just need the technical methods to extract them efficiently. If you're comfortable with spreadsheets and basic statistics, you have enough background to start applying supervised learning effectively.

We don't teach general programming or explain what finance is. We assume you're already analyzing data and want better tools for prediction, classification, and pattern recognition. Most participants work at banks, insurance companies, investment firms, or fintech startups in South Africa. They come with specific questions about their current projects and leave with working code they can modify for their needs.