Image classification using machine learning algorithms is a rapidly growing area of computer science. This project explores this area by using a convolutional neural network to classify backyard birds and examines an end-to-end solution to this task by deploying the network after training. Specifically, this involved generating a dataset with more than 12,000 images, labelling that dataset, and then using it to train the neural network using supervised learning. To improve performance of the network, its various hyperparameters were tuned over the course of many experiments and data was augmented to generalize the learned features. Finally, the trained convolutional neural network was deployed in a custom Android application.