Trade-tested, MQA-certified electrician | BSc Geology & Chemistry | PGDip Data Science (Wits, completing Dec 2026)
I apply Python, SQL and Power BI to industrial problems, drawing on three years inside a platinum smelter and a science background in the materials being processed. Open to data analytics and modelling roles in mining, metals and heavy industry.
📍 Polokwane / Johannesburg, South Africa 📫 mmama.mosima@gmail.com 🔗 LinkedIn
- Electrician: completed a three-year apprenticeship at the Anglo American Polokwane Smelter (Valterra Platinum) and qualified as a trade-tested, MQA-certified electrician. I know plant equipment, maintenance cycles, Zero Harm practice and MHSA requirements first-hand.
- Scientist: BSc in Geology and Chemistry (University of Limpopo, 2016), so I understand the material being mined and processed.
- Data scientist in training: Postgraduate Diploma in Data Science at the University of the Witwatersrand (completing December 2026), with an MSc in Data Science planned from 2027. Coursework covers machine learning, statistical and spatial modelling, and data validation.
| Area | Tools and methods |
|---|---|
| Programming and data | Python (Pandas, NumPy, Matplotlib, Seaborn, Plotly), SQL, Pytest, Git |
| Dashboards and apps | Power BI, Streamlit |
| Modelling | Statistical modelling, spatial prediction, machine learning, anomaly detection |
| Business systems | SAP material master, B670 requisitions |
Industrial compliance automation | View repo
| Tech | Python, Pandas, NumPy, Git |
| What it does | Automates electrical installation checks against SANS 10142-1: calculates cable voltage drop and flags non-compliant circuits. |
| Why it matters | Replaces manual Excel checks and supports asset compliance and safety audits in mining and smelter environments. |
| Includes | Sample cable data and the SANS 10142-1 formulas. Built by a qualified electrician. |
Data cleaning and metrics pipeline | View repo
| Tech | Python, SQL, Pytest |
| What it does | Cleans and validates workforce data (nulls, text standardisation, date ranges 2015–2025 with leap-year logic) and calculates retention rate, turnover and averages. |
| Context | Built for the Wits PGDip Programming for Data Science module on a course-supplied dataset, which is not included in the repository. |
| Transferable to | The same cleaning and validation approach applies to maintenance and equipment records. |
Data analytics and modelling roles where engineering practice and data methods meet: equipment reliability, maintenance analytics and operational efficiency in mining, metals and heavy industry.