I developed a deep learning project in image classification, training a model in Tensorflow on the CIFAR-10 dataset. By building and training convolutional neural networks, I explored different techniques for optimization and model architectures in order to improve accuracy. I began by defining the problem, selecting a metric to measure success, determining the evaluation protocol, preparing the data, and developing a model that outperforms the baseline. I then scaled the model, observed overfitting, and applied regularization techniques while tuning the hyperparameters. Throughout the process, I documented my methods and findings in a Jupyter notebook, linked to the left. This notebook provides a clear report of my experiments, results, and insights gained.