I am a second-year Computer Science PhD student at Cornell Tech, gratefully advised by Prof. Vitaly Shmatikov. Broadly, my interests are somewhere in the (vast) union of machine learning, robustness, security, privacy, and fairness. Nowadays, I'm mainly thinking about the vulnerabilities baked into modern encoders and embedding spaces.
I recently completed my Master's degree in Computer Science at the University of Washington, Seattle (UW), where I was advised by Prof. Sewoong Oh. My thesis, Label Poisoning is All You Need proposes a novel trajectory-matching-based backdoor attack, that corrupts (i.e., 'poisons') only the labels in a training set to create a highly-effective backdoor with an arbitrary trigger.
Before that, I received a B.S.B.A. in Computer Science and Mathematics, Philosophy with cum laude and ΦΒΚ honors from UW. During my stay, I was advised by Prof. Sewoong Oh, Prof. Rajesh Rao, and Spencer Sevilla. Additionally, I spent five summers working at Microsoft's Security and AI research group, building machine-learning detections for large-scale criminally-orchestrated attacks.
You can reach me via the Connect tab on the main page.