John Kirchenbauer is a Postdoctoral Fellow at the Vector Institute, where he works with Prof. Colin Raffel on fully open language model development and mitigating the risks posed by modern AI. He completed his PhD at the University of Maryland with Prof. Tom Goldstein in 2026, studying methods for enhancing trust and transparency throughout the language-model development and deployment lifecycle. More broadly, his research spans robustness, reliability, safety, and scalability in deep learning, unified by a long-standing interest in how training data shapes generative-model behavior. During his PhD, he interned at Google X; before that, he was a research engineer at Carnegie Mellon University’s Software Engineering Institute. He holds BS and MS degrees in Computer Science from Washington University in St. Louis (2020) and a Performance Diploma in Violin from the Oberlin Conservatory of Music (2017).