We will describe the training and development of convolutional neural networks for protein-ligand scoring and how these deep learning models are integrated into the GNINA molecular docking open source software. Successful prospective evaluations of GNINA will be discussed, including recent top performance in the Critical Assessment of Computational Hit-Finding Experiments (CACHE). Additionally, we will describe our open source pharmacophore screening resource, Pharmit, which enables the screening of millions of compounds in seconds and discuss several generative approaches for hit discovery using deep generative models.