Foad Namjoo

I'm a Computer Science PhD student at the Kahlert School of Computing, University of Utah, advised by Professor Jeff Phillips in the UtahDB Lab — PhD expected May 2028. My research makes LLMs more interpretable — using activation steering to direct model behavior such as truthfulness and verbosity, in collaboration with Martian AI. On the theory side, I develop distance metrics for high-dimensional distributions — dKS extends the Kolmogorov–Smirnov distance to multiple dimensions as a true metric, with near-linear algorithms (~76,000× faster than baseline at 1M samples) shipped as an open-source C++/Python library. And on the systems side, MotionPI, a wearable sensing platform, powers an NIH-funded field study at ~7.7M records a day with zero malformed writes.