This project focuses on classifying handwritten digits (0–9) using the MNIST dataset with a Deep Learning model. The dataset consists of 28x28 pixel grayscale images, which were preprocessed and fed into a neural network. Through this project, I gained hands-on experience with: 1)Deep Learning architectures ,building and understanding the structure of neural networks. 2)Activation functions ,exploring how functions like ReLU and Softmax impact learning and predictions. 3)Model training and evaluation ,splitting data, training the model, and testing accuracy 4)Image data representation ,understanding how pixel values are transformed into meaningful features for the model. The final model was able to accurately recognize handwritten digits, reinforcing my understanding of how Deep Learning models process and classify images. The model achieved 97% accuracy on test data.