My research program is focused on developing an improved understanding of earthquakes, fault zones, and the hazards they pose to society. Large earthquakes are very difficult to study in detail because they are infrequent, with typical time between repeat events being hundreds to thousands of years. Instead, I aim to better understand the physics of earthquakes and faults by studying the interactions of millions of tiny earthquakes that are occurring nearly all the time. My group pioneered the use of machine learning techniques to better identify these tiny earthquakes which has led to nearly an order of magnitude more events being discovered. We have used these catalogs of tiny earthquakes to image fault zones in unprecedented resolution and resolve the dynamic physical processes that drive earthquake sequences. Recently, we have discovered a new type of earthquake swarm in California that is responsible for a substantial amount of the daily earthquake activity.