De novo drug design (DNDD) uses a computational approach to allow researchers to design proprietary chemical entities, compounds, and drug candidates more quickly and economically than other drug design methods. While conventional DNDD methodologies use only the information regarding a biological target and its known active binders, recent advances in de novo discovery are leveraging quantum mechanical calculations as well as machine learning (ML), deep learning (DL), and other related AI-based technologies to explore the broadest chemical spaces possible – utilizing both state of the art computational chemistry methods and vast data troves from available scientific literature and databases. So while standard de novo design methods can be limited by existing knowledge around receptor–ligand interactions, breakthrough deep learning methods combine advancements in computational chemistry and artificial neural networks with reinforcement learning in order to discover novel drug candidates that are virtually “template-free.”