Utilized BERT and RoBERTa models for sentiment classification and tokenization of tweets. Customized a PyTorch dataset class and implemented a training-testing split of 90%-10%.

Developed an interpretable AI model for churn causation and retention strategies. Achieved 91% accuracy in predicting customer churn over a 5-week period. Predicted sixth week's sales using machine learning, achieving the lowest RMSE.

Developed a predictive ensemble ML model using Sklearn, Seaborn, XGBoost, LightGBM, CatBoost and others. Targeted patient readmission within 90 days due to heart failure using a structured dataset. Executed data preparation, various analyses, and oversampling.

Consistently maintained high academic performance throughout the degree program, earning recognition on the Dean's List for three consecutive terms.

Awarded Bronze Medal for outstanding academic achievement in the BS Artificial Intelligence program.