Mfundo Monchwe
Data Scientist | Analytics Engineer | Statistician
Turning complex data into practical decisions
I am a Data Scientist, Analytics Engineer, and Statistician with 4+ years of experience working across finance, healthcare, retail, and mining.
My work sits at the intersection of statistical modelling, machine learning, analytics engineering, data quality, and decision support. I enjoy taking imperfect real-world data, understanding the underlying problem, building an appropriate analytical solution, and translating the result into something people can actually use.
I hold an MSc in Statistics (Cum Laude) from the University of KwaZulu-Natal and have experience spanning pricing analytics, forecasting, credit risk, customer segmentation, dashboards, APIs, bioinformatics, and applied machine learning.
At a glance
4+ years
Applied data science and analytics experience
MSc Statistics
Cum Laude, University of KwaZulu-Natal
2 publications
Peer-reviewed machine learning and computer science research
Core stack
Python · pandas · NumPy · SQL · R · Scikit-learn
What I work on
Analytics & statistical modelling
Exploratory data analysis, data validation, hypothesis testing, regression, time-series forecasting, customer analytics, pricing analytics, and business reporting.
Machine learning & deployment
Predictive modelling, classification, credit risk, model interpretability, dashboards, REST APIs, and lightweight production-facing analytical applications.
Modern AI engineering
Hands-on learning and implementation across Generative AI, LLM applications, Retrieval-Augmented Generation (RAG), local and API-based models, tool calling, and agentic AI.
Featured work
Vehicle pricing, forecasting, and customer analytics
At Avis Budget, I worked across pricing and analytical use cases including vehicle price prediction, ARIMA-based termination and maintenance forecasting, customer profiling and segmentation, loss-ratio analysis, credit-risk recommendations, and multi-country data-quality analysis.
Credit scoring and model delivery
I designed and commercialised a credit-scoring solution exposed through a REST API, incorporated Explainable AI, implemented secure access using Keycloak, and developed a Streamlit interface for real-time model interaction.
Research in probabilistic machine learning
My published work includes Bayesian convolutional neural networks for COVID-19 lung-image classification with uncertainty estimation, alongside research in digital-signature security.
Currently developing
I maintain a hands-on learning repository focused on modern AI engineering. The goal is to understand the systems behind contemporary AI applications rather than only use high-level tools.
Current areas include:
- Generative AI and large language models
- API-based and local-model workflows
- Retrieval-Augmented Generation (RAG)
- Knowledge-grounded responses
- Tool calling
- Agentic AI and multi-step workflows
Let’s connect
I am interested in opportunities and collaborations across data science, analytics engineering, statistical modelling, machine learning, and applied AI.