Mirac Suzgun is a PhD student in Computer Science at Stanford University, advised by Dan Jurafsky, and a JD candidate at Stanford Law School. As part of his doctoral studies, he has been studying the limitations and capabilities of large language models and exploring more efficient and interpretable methods to perform text generation. He previously completed his undergraduate education at Harvard College, pursuing a joint degree in Mathematics and Computer Science and a secondary in Folklore & Mythology. His current work at RegLab includes profiling legal hallucinations in large language models and using machine learning tools to identify racially restrictive covenants in property deeds.