Before starting at MIT, I spent 2 years doing research at the Allen Institute for AI (AI2) and also spent a summer interning at Microsoft Research. I graduated from Harvard University in 2020, where I studied Computer Science & Philosophy (joint degree).
Research Interests These days, I spend most of my time thinking about what it would take to build personalized AI tutors that can adapt to individual students. Achieving this goal requires being able to model student reasoning (including any misconceptions): Two recent projects took steps in this direction in the math and coding domains.
Please feel free to get in touch if you're also thinking about personalization, education, or user/student modeling.
News
| Oct 2025 | Two new preprints on modeling student reasoning in coding and math. |
| Sept 2024 | Invited speaker at the Harvard CMSA Panel on Machine Learning in Science Education. |
| Sept 2024 | Invited talk to the Google Deepmind LearnLM Team. |
| June 2024 | New preprint on language modeling with editable external knowledge. |
| May 2024 | Two papers accepted at ACL 2024: adaptive teaching toward misconceptions and editing scientific papers in response to reviews. |
| March 2024 | Paper on counterfactual evaluations of LLMs to be presented at NAACL 2024. |
| June 2023 | Started at the Diverse Intelligences Summer Institute. |
| May 2023 | Paper on rationalization & counterfactuals accepted at ACL 2023. |
| April 2023 | Gave an invited talk at the IST & Unbabel Seminars. |
| April 2023 | Selected for the NSF GRFP fellowship. |
| Jan 2023 | Selected as a winner of the inverse scaling prize. |
| Dec 2022 | Presented our paper on self-rationalization & robustness at EMNLP 2022. |
| Oct 2022 | We've released cs-sop.org, a repo of example CS SOPs for grad school applications! |