Hi! I am a PhD student in the Computer Science department at Stanford University. I am co-advised by Dorsa Sadigh and Dan Boneh. I am also an AI Science Writer at Anthropic. My research focus is on building AI systems that will positively impact student learning and education, which includes:

• Student Modeling: I study how to accurately model student learning trajectories (e.g. mistakes, skill acquisition) on platforms such as foreign language learning on Duolingo and introductory programming on Pencil Code, as well as the effects of AI assistance on knowledge and reasoning tasks.
• Algorithmic Teaching and Discovery: As we develop black-box AI models to discover solutions to complex tasks, what algorithms can help teach humans the underlying strategies that are learned? I explore the use of skill discovery algorithms and shared autonomy for motor learning, as well as language models for feedback generation, interactive control, and cultural transmission.
• Safe and Secure Deployment: What security and robustness concerns arise when deploying AI systems in high-stakes domains? My work has shown that users are more likely to introduce security vulnerabilities with code generation models, and I am also excited about technical problems such as controlling GPU nondeterminism for verifiable training and mitigating bias amplification in online learning.

During my PhD, I spent time visiting MIT LINGO in Cambridge, MA. Previously, I was an AI Resident at Microsoft Research, and also worked with Hoda Heidari and Andreas Krause on algorithmic fairness through the amazing ETH Zürich Summer Research Fellowship. I completed my undergrad in Computer Science and Creative Writing at Stanford, where I was a member of the Stanford NLP group, and worked with Tatsu Hashimoto, Percy Liang, and Kalanit Grill-Spector.

Updates