Potato is a significant crop for its nutritional value, tonnage capacity, and profit potential. It is a very highly consumed people’s staple crop, whose intake is on the upward push due to the global pandemic. Misidentification and late detection of diseases in potatoes further deteriorate the plant condition, resulting in severe risks to the quality and yield of harvests. Early blight and late blight are the most common diseases that attack potato plants, and they seriously lead to losses in the economy of most farmers. This work involves a complete implementation of a digital potato disease detection system based on our developed robust deep learning Convolutional Neural Network (CNN) model. The CNN model was developed using 2152 dataset images taken from Kaggle. These images are categorized into three groups: early blight, late blight, and healthy leaves. The dataset is divided into three subsets: training (80%), validation (10%), and testing (10%). We constructed a robust disease detection model using Convolutional Neural Network (CNN) algorithms. The model achieved an impressive accuracy of 98.21% by learning from the underlying patterns in raw images. This interactive platform empowers farmers to detect and manage potato crop diseases efficiently, thereby enhancing agricultural productivity. During our discussion, we also covered performance measures.