Induced pluripotent stem cell derived cardiomyocytes (iPSC-CMs) provide an opportunity to study cardiac development, cardiac disease, and cardiotoxicity. Here, we present a deep learning model that has been trained on a 100,000 image dataset of induced pluripotent stem cells (iPSCs) differentiating into cardiomyocytes. We demonstrate that the model is able to predict differentiation outcomes through classification and segmentation. We then investigate the molecular mechanism underlying the model’s predictions and finally demonstrate that the model can be used for high-throughput label-free screening of drugs and environmental toxins that can perturb cardiac development.