who am i?

professionally speaking...

I'm Atrey Desai, a third-year undergraduate student studying computer science and linguistics with a minor in korean studies at the University of Maryland. I am fortunate to be advised by Professor Rachel Rudinger and Professor Jordan Boyd-Graber.

I am also eternally thankful for Nishant Balepur, Yoo Yeon Sung, and Sathvik Nair for taking a curious undergraduate under their wing.

I am a member of the Computational Linguistics and Information Processing (CLIP) lab in UMIACS. Currently, I'm also a visiting researcher under Professor Kenny Zhu at the Arlington Computational Linguistics Lab (ACL2) at the University of Texas at Arlington. Our work focuses on identifying structural similarities to human language and contextual semantics within animal vocalizations to gain insights about the language of animal species.[1]

I was previously a member of the technical staff of Learn Prompting, where I worked on HackAPrompt and research on creating trustworthy and robust AI safety judges.[2]

Previously, I worked under Professor Michael Littman at the Reinforcement Learning & Adaptive Behavior (RLAB) group at Brown University on applications of reinforcement learning to 2D non-sequential tasks.[3]

personally speaking...

I would describe myself as curious and bubbly. I love exploring new ideas and live at the boundary between disciplines, especially when they reveal unexpected structure. Naturally, this impulse led me to NLP, a field where computer science meets reading and journalism, my favorite pastimes growing up.

On the day to day, you'll probably find me at the lab, or at the gym, or at Moge Tea, or wandering about aimlessly looking through my camera viewfinder!

other interests: photography, reading[4], graphic design[5], learning languages, punching the air, music (especially edm and indie pop), manifold[6], and cultures of the world


where are you from?

I was born and grew up in Saratoga, California, attending the lovely Saratoga High School. I currently live in the Washington, DC area, specifically in College Park, MD. I also tend to frequent the New York City and Boston areas for work and visiting friends. If you are ever around and want to talk, feel free to reach out!


why does this website look like this?

inspiration: This site draws inspiration from many beautiful personal websites that I had been collecting for the better half of 2025.

  • Aryaman AroraFor the click to reveal email feature, formatting for the about section, and footnotes.
  • Henry WangFor cool effects when hovering over the profile photo.
  • Preston FuFor providing a rough base for the website, formatting content in 1:3:1, and hovering over research images for animated previews.
  • Jessy LinFor inspiring the formatting for the photography section.
  • Brian LovinFor collecting a host of sites to collect tidbits of animations from.
  • Lingxi LiFor the current time in the footer.
  • Maxime HeckelFor general creative inspiration and a better implementation of the lego effect.
  • Mason WangFor the design and thought process behind the bookshelf section.
  • Lily YangFor raise for hover section animation inspiration, and the music in the footer.
  • Are.naFor general website inspo.
  • ShashwatFor inspiring a custom cursor.
  • Christy LiFor the hover and shadow animations on highlighted cards

built with: This website was built in Antigravity, with assistance from GPT 5.4, Opus 4.5 and Gemini 3 Flash. It uses SvelteKit, Tailwind CSS, and is deployed on Vercel.


things i'm thinking about lately

  • what to focus on in my phd
  • things to photograph in my daily life
  • why are sparkling ice drinks so good
  • cheesecake
  • how does the weather change from mid-80s to snow in 1 day
  • key lime pie

footnotes

1. This work at ACL2 aims to eventually create an end-to-end machine translation pipeline for animal vocalizations. Our research is also graciously supported by the NSF.
2. HackAPrompt is the world's largest red-teaming hackathon.
3. Our findings were published at the AAAI-22 Workshop on Interactive Machine Learning (IMLW) and the Conference on Reinforcement Learning and Decision Making (RLDM-22). And also my first publications!
4. I especially love collecting used and old books. I tend to buy too many from the discount carts outside the Strand in NYC.
5. A relic of high school newspaper and my wonderful journalism teacher, Mr. Tyler. I had also been an editor for STYLUS, Maryland's literary arts and photography magazine my freshman year.
6. This had started out as a fun thing with my roommate, but we both discovered we were really good at predictions. It became a nightly tradition to come back to the apartment and talk about the progress of our trades for the day (much to the chagrin of our friends).
Washington, DC
Last updated Mar 24, 2026