With the rapid growth of e-commerce platforms, customers frequently share their opinions about products through online reviews. These reviews provide valuable insights into product quality, delivery services, and overall customer satisfaction. However, the large volume of user-generated content makes manual analysis difficult. Sentiment analysis, a Natural Language Processing (NLP) technique, helps automatically identify the emotional tone behind textual data. This paper presents a sentiment analysis system designed to classify Flipkart product reviews into positive, negative, and neutral categories using the DistilBERT model. DistilBERT is a lightweight transformer-based model derived from BERT that provides strong contextual language understanding with reduced computational cost. The proposed system includes stages such as data preprocessing, tokenization, model training, evaluation, and visualization. The model is evaluated using performance metrics such as accuracy, precision, recall, and F1-score. Experimental results demonstrate that the DistilBERT model effectively captures contextual relationships between words and produces reliable sentiment predictions. The developed system can assist businesses in analyzing customer feedback efficiently and support data-driven decision-making in e-commerce platforms.