Theresa Sawi, U.S. Geological Survey Repeating earthquakes sequences are widespread along California’s San Andreas fault (SAF) system and are vital for studying earthquake source processes, fault properties, and improving seismic hazard models. In this talk, I’ll be discussing an unsupervised machine learning‐based method for detecting repeating earthquake sequences (RES) to expand existing RES catalogs or to perform initial, exploratory searches. This method reduces spectrograms of earthquake waveforms into low-dimensionality “fingerprints” that can then be clustered into similar groups independent of initial earthquake locations, allowing for a global search of similar earthquakes whose locations can afterwards be precisely determined via double-difference relocation. We apply this method to ∼4000 small (⁠Ml 0–3.5) located on a 10-km-long creeping segment of SAF and double the number of detected RES, allowing for greater spatial coverage of slip‐rate estimations at seismogenic depths. This method is complimentary to existing cross‐correlation‐based methods, leading to more complete RES catalogs and a better understanding of slip rates at depth.