Maha Naeem

Delivered applied machine learning solutions for international clients across insurance, finance, legal-tech, retail, and e-commerce domains. Worked independently across the full ML lifecycle — from raw data ingestion to deployment-ready APIs. Selected Projects: Customer Churn Prediction Built classification models on a 120k-row dataset using logistic regression and XGBoost. Applied SMOTE to handle imbalance and engineered behavioural features from policy history. Improved ROC-AUC from 0.71 to 0.84.

Developed and optimised dynamic insurance application forms using JavaScript and structured validation logic. Improved form completion rates by simplifying multi-step flows and reducing drop-offs. This role laid the foundation for my structured logic and client-focused product thinking — skills I later applied to ML solution design.

Contributed to desktop and web application development using MVC architecture. Worked on back-end logic and database integration alongside senior engineers, gaining exposure to production-grade code organisation and deployment workflows.