I’m a PhD student who works on developing and applying machine learning methods to study human behavior and design tools to help us communicate more effectively. Some recent projects I’ve worked on have included a large-scale study of polarization on Reddit and Twitter, a new way of jointly modeling graph and text embeddings through contrastive pretraining, and methods to identify “bridging” influencers on social networks that can connect disparate groups. I am also broadly interested in how we can align language models well with different communities. If you are interested in NLP, computational social science, or experimental design feel free to reach out!