About
I'm a Master's student in the Machine Learning Department at Carnegie Mellon University, where I work with Prof. Albert Gu. My current research explores dynamic tokenization for time-series forecasting using H-Nets. I've also worked with Prof. Beidi Chen on long-horizon multimodal memory systems for agents.
Before CMU, I was a Software Engineer at Google, working across the Cloud and Assistant NLP teams. Alongside my engineering work, I pursued research with Google Research on time-series foundation models, scaling decomposition-based approaches to billion-parameter models, designing training data mixtures, and developing multi-agent systems to improve forecasting performance. I've also interned at IST Austria with Prof. Dan Alistarh's lab, where I worked on generating automatic reviews for ML conference papers.
My broader research interests include foundation models, efficient sequence modeling, and representation learning.
Publications
-
Decriminalizing Same-Sex Relationships and Its Impact on Social Media Discussions in Asia
-
The MediaSpin Dataset: Post-Publication News Headline Edits Annotated for Media Bias
-
FLAIRR-TS: Forecasting LLM-Agents with Iterative Refinement and Retrieval for Time Series
-
"Reasoning" with Rhetoric: On the Style-Evidence Tradeoff in LLM-Generated Counter-Arguments
-
Towards Foundation Models for Time Series Forecasting
Beyond Research
Outside of research, I enjoy exploring local coffee shops and painting. I'm always happy to chat about research ideas, so feel free to reach out, especially if we've met before!