Presenting two papers at ACL 2025: one on an interview study characterizing the sociotechnical gap between what experts in materials science and law/policy do, and where NLP research is focused; the other introducing Collage, a tool for facilitating rapid prototyping, co-design, and debugging of information extraction approaches on PDFs. Also presenting work on data-driven materials design as a new benchmark for information extraction at the AI and Scientific Discovery Workshop at NAACL 2025.
Research Experience
2015-2018: Worked at IBM Watson on a team that did bespoke prototypes; 2018-2021: At Kensho Technologies as an ML engineer focusing on NLP, then became the first lead of the ML Ops and Internal Tools team.
Education
PhD: Language Technologies Institute at Carnegie Mellon University, advised by Emma Strubell; Master's degree: Carnegie Mellon University, supervised by Carolyn Rose; BA: Computer Science from Columbia University.
Background
Research interests: NLP and AI tools that allow users in specialized domains to maintain agency in their work. Specifically, focuses on user-customizable, on-device models that live in the browser, and how to effectively reason about their limitations and update them. Also interested in the incentives that shape NLP research, whether funding, tooling, or culture.
Miscellany
Planning to intern at Ikigai Labs, exploring the intersection of UX and ML, working on making user-controllable models for timeseries prediction.