Food poisoning is a problem affecting people on a global scale, killing 420,000 people a year as of 2022. This problem is exacerbated by the distribution of already-rotten food from farms to vendors and can be mitigated by preventing infected food from ever leaving farms in the first place, or by identifying rotten foods before vendors sell them. Thus, this study summarizes the building of a hyperparameter-optimized deep learning model that uses a Convolutional Neural Network (CNN) to identify the kind of fruit and its quality by looking at an image of a fruit, a simple process. This automation of food classification and safety allows lower-income farmers and vendors to escape the time and monetary cost of manually verifying whether or not each fruit they distribute/sell is safe to eat.