In today’s fast-paced e-commerce environment, efficient and accurate billing processes are essential to maintaining customer satisfaction and reducing operational costs. Traditional barcode scanning can be labour-intensive, prone to errors, and dependent on manual intervention. This project introduces an innovative machine learning (ML)-based approach that leverages computer vision to streamline the billing process in a grocery e-commerce setting. By utilizing the YOLO v7 object detection model, this system automates the identification and billing of items without requiring barcodes. When a customer places a grocery order, the order details are relayed to the warehouse. There, instead of scanning barcodes, a camera equipped with the YOLO v7-based ML model captures images of the selected items, automatically detecting and identifying each product. The algorithm then generates a bill based on the detected items and compares it to the customer's original order list. This comparison enables the system to validate whether the detected items match the order list and provides feedback on any discrepancies—highlighting missing or extra items if there is a mismatch. The model is designed to recognize up to 30 different product classes, with the potential for future scalability to accommodate additional products. This solution will not only reduce the dependency on barcode systems but also enhances accuracy, reduces human error, and improves operational efficiency. This approach can be adapted for diverse product lines, offering a robust alternative for automated item verification and billing in e-commerce and other retail settings.