About

I am a Computer Science graduate student at the University of Arizona, working with Mihai Surdeanu and Ellen Riloff. My research focuses on measuring the capabilities, reliability, and faithfulness of AI systems, while finding ways to make them better collaborators and assistants.

My research journey began as an undergraduate, driven by both the potential and the failures of LLMs, particularly hallucinations. With Steven Bethard, I worked on authorship-attribution approaches to machine-generated text detection; with Mihai Surdeanu and Eduardo Blanco, I explored the limits of LLM reasoning in non-linear debate settings.

More recently, I’ve been thinking about personal assistant agents: how we should measure their productivity, how they can improve over time, and how human-agent interaction can move beyond the traditional chat interface.