Grape cultivation is vital to the global economy, but various diseases threaten grapevine health and yield. Early and accurate disease detection is crucial for implementing effective control measures. Traditional methods rely on visual inspection by experts, which is subjective and time-consuming. Automated approaches using deep learning offer a promising alternative. The goal of this study was to create a convolutional neural network (CNN) model that can identify grapevine leaf diseases reliably. The aim of the study was to train the model for distinguishing between healthy and infected leaves by three diseases that are common in them such as black rot, esca, and leaf blight. We collected a dataset of images containing healthy and diseased grape leaves. The dataset was pre-processed and augmented to increase data diversity and ensure consistency. The CNN model was developed using TensorFlow and Keras and consisted of convolutional, pooling, dense, and activation layers. Optimizations were implemented using cross-entropy loss function, adam optimizer, and Dropout layers. Model performance was evaluated using metrics such as accuracy of the model, the precision at which it cannot wrongly label a negative sample, the recall at which it can find all the positive outputs, and F1-score which is the ratio between precision and recall. The developed CNN model achieved high accuracy in disease identification. We observed significant accuracy, precision, F1-score, and recall values across all disease categories, demonstrating the model's effectiveness in distinguishing between healthy and diseased leaves. This study successfully demonstrates the potential of CNNs for automated grape leaf disease identification. The developed model offers a promising tool for viticulturists to improve early disease detection and ultimately enhance grape yield and quality. Future research could focus on expanding the model's capabilities to identify a wider range of diseases and exploring its potential for real-time applications in vineyards.