Machine learning can be used to screen billions of molecules for potential use as therapeutic drugs. In pursuit of utilizing the current state-of-the-art ML models, companies have invested enormous sums of money into massive computing infrastructure with large GPU footprints. Is this really necessary? Here we introduce the Tsetlin Machine, a low-resource, continuous and interpretable ML method which aims to provide a high capacity of drug design for a fraction of a fraction of the computational cost of existing Neural Network methods.