I turn messy datasets into decisions - and I explain them so anyone can follow.

I'm Vusumuzi Nkosi, a data scientist with postgraduate training in Data Science (STADIO, NQF 8) and 9+ years as a mathematics lecturer. That teaching background isn't a footnote - it's the thing that makes me different. I don't just build models; I communicate what they mean.

I hold a PGDip in TVET from the University of the Western Cape and am an ALX Data Science graduate. I work at the intersection of data science and finance, building end-to-end pipelines on real-world data.

• Education: PGDip Data Science @ STADIO (NQF 8) | PGDip TVET @ UWC | ALX DS Graduate | BSc Maths & CS @ Unisa
• Location: Newcastle, KZN, South Africa
• Focus: Credit Risk | Financial Analytics | Machine Learning
• Open to: Data Scientist, Data Analyst, Credit Risk Analyst roles (SA / remote)

2.26M real loans | AUC 0.72 | ~$95M annual impact

End-to-end credit-risk pipeline on real Lending Club data (2007-2018). SQL schema design, feature reduction from 151 to 33 variables, full EDA, Gradient Boosting model evaluation, and ~$95M/year in quantified preventable losses. Includes Excel dashboards and a deployed Streamlit prediction app.

Tech:

World Bank data | Python ETL | SQL + Power BI + Excel

Financial indicators dashboard built from World Bank data on South African economic metrics. Python ETL scripts for data collection and cleaning, SQL for structured storage, and Excel/Power BI reporting views.

Tech: