FinSenseAI is an intelligent financial transaction categorization system designed to automate the classification of banking transactions into meaningful spending categories. The project combines Natural Language Processing (NLP) with machine learning to reduce manual effort and improve the accuracy of financial data organization. I fine-tuned a DistilBERT model to understand transaction descriptions and built a complete end-to-end pipeline for data preprocessing, model training, evaluation, and deployment. The application includes an interactive Streamlit interface that supports both individual transaction prediction and bulk CSV processing, along with visual analytics to help users understand their spending patterns. Key Results: Fine-tuned a DistilBERT model on a dataset of 12,000+ financial transaction records. Achieved 97% classification accuracy in categorizing transaction descriptions across multiple spending categories. Developed a Streamlit application supporting both single transaction prediction and bulk CSV classification. Built a complete NLP pipeline covering data preprocessing, feature preparation, model training, evaluation, and deployment. Improved transaction organization by significantly reducing the manual effort required for financial record categorization.