My presentation introduces an AI model to detect and accurately classify handwritten digits from 0 to 9. The model utilizes a neural network architecture trained on a large dataset of handwritten numerals to achieve high accuracy in digit recognition. The model is optimized to consider handwriting style deviations and noise through preprocessing methods like normalization and data augmentation. The results present great efficiency, with the model performing near-human accuracy on standard datasets such as MNIST. This project displays possibilities for deep learning in practical applications such as digit recognition for forms, checks, and automated data entry systems.