With the avalanche of alerts to be delivered by Rubin Observatory's Legacy Survey of Space and Time and the limited resources for follow-up, we will need brokers to select intriguing alerts that warrant follow-up in a timely manner. At NSF's NOIRLab and University of Arizona, we are developing the Arizona-NOIRLab Temporal Analysis and Response to Events System (ANTARES, Saha et al. 2014, 2016, Narayan et al. 2018), to hunt for the rarest of the rare events in the time domain. In this work, we provide an overview of the ANTARES system, how we use real-time alerts from the ongoing Zwicky Transient Facility survey as a training set, and the way forwards to Rubin observatory.