The Vera C. Rubin Observatory Legacy Survey of Space and Time (LSST) will produce 10 million transient notifications per night and will observe 100,000s of supernovae over 10 years. Spectroscopic resources will not be able to keep up with the demand for classifications of these objects. In this talk, I will present ways that the Rubin Observatory LSST Dark Energy Science Collaboration (DESC) is preparing for large quantities of unclassified data using machine learning. I will overview the Photometric LSST Astronomical Time-series Classification Challenge (PLAsTiCC) aimed at engaging the global data science and machine learning community to develop new methods for classifying an LSST-like survey and present solutions to this challenge. Additionally, I will introduce the Recommendation System for Spectroscopic follow-up (RESSPECT) that is specifically geared towards improved supernova cosmology with a photometrically classified sample. We use active learning to account for multiple metrics to create a targeted follow-up plan to send to observers every 24-hours.