Yanga Gcabo

At UNISA, I developed a Peak Sales Forecasting System to predict high-demand periods, promotional spikes, and seasonal fluctuations. I combined statistical forecasting, machine-learning models, and deep-learning architectures in a pipeline with time-series cross-validation and feature engineering. I designed the system to be deployable through a FastAPI microservice and Docker container, with an optional Streamlit dashboard. My work draws on the data engineering experience I gained during training at KFC, where real-time operational constraints shaped my understanding of production-ready data systems. In my Integrated Engineering Project at CPUT, my group designed and built a davit crane capable of lifting up to 1700 kg. I handled mathematical modelling, load calculations, stress analysis, and control simulations, and programmed the hoist and safety logic using PLC Structured Text and Function Block Diagram.