Data Science / AI Intern at Y2S

Contributed to the development of a customer churn prediction pipeline in Python (Pandas, Scikit-Learn, XGBoost) on transactional data, supporting segmentation of high-risk customers for retention analysis. Optimised data preprocessing workflows including imputation, deduplication, and anomaly detection to improve dataset quality for downstream modelling. Participated in model training and benchmarking across Logistic Regression, Random Forest, and XGBoost; the best model achieved ROC-AUC of 0.86 and F1-score of 0.78.