Vatsal Baherwani

about me

Hey there, I'm Vatsal, a PhD student currently living in New York. I work with Pavel Izmailov and Andrew Gordon Wilson at NYU, and I am also grateful to be supported by the NSF Graduate Research Fellowship.

This past summer I interned at Q Labs in San Francisco, where I researched scalable methods for synthetic data generation, specifically for language models. The summer before that, I was a visiting researcher at the Center for Human-Compatible AI at UC Berkeley, working with Raj Movva and Emma Pierson on practical applications for ML interpretability.

I previously graduated with a bachelor's degree from the University of Maryland, where I was fortunate to work with Pete Kyle (who first introduced me to academic research), Abhinav Shrivastava, Ashwinee Panda, and Tom Goldstein.

research

I am broadly interested in developing a scientific understanding of deep learning; some specific topics I have been exploring recently are emergent capabilities in language models, applying AI for scientific discovery, and generating synthetic data for pretraining.

selected publications

more about me

• I used to write daily blog posts, although I don't as often anymore because I am trying to transition into running a social media vlogging page instead.
• I love food. If you can recommend a good restaurant in NYC, please contact me. I will gladly return the favor.
• Recently, I have been learning to sing and dance (specifically Bollywood dance). If you know of any fun dance classes/events in NYC let me know!
• I really enjoy reading about human nature, the brain, and consciousness. Here's what my bookshelf looks like.
• I want to travel as much of the world as possible in my lifetime. So far I've only been to a handful of countries.

links

• Email
• Google Scholar
• Twitter
• GitHub
• LinkedIn