I’m Amitayush Thakur, a Ph.D. student at UT Austin advised by Prof. Swarat Chaudhuri.
My research interest is broadly at the intersection of “AI for Mathematics” and “AI for Code”. I’m interested in automated mathematical reasoning through LLMs and its implication in fully verified program synthesis essential for generating industry-grade code from AI. I’m also interested in program synthesis from natural language, neuro-symbolic AI, explainability and interpretability of Deep Neural Nets through symbols.
Before joining UT, I worked as Software Engineer 2 at Microsoft in Azure, and before that as a research intern at Microsoft Research.
I did my Bachelors in Computer Science and Masters in Mathematics at BITS Pilani.
Besides that, I also enjoy solving fun mathematics problems, reading about number theory and cryptography.
Recent Preprints and Publications
| Thakur, Amitayush, George Tsoukalas, Yeming Wen, Jimmy Xin, and Swarat Chaudhuri. "An In-Context Learning Agent for Formal Theorem-Proving." First Conference on Language Modeling (COLM), 2024. [arXiv], [Code] |  |
| Tsoukalas, George, Jasper Lee, John Jennings, Jimmy Xin, Michelle Ding, Michael Jennings, Amitayush Thakur, and Swarat Chaudhuri."PutnamBench: Evaluating Neural Theorem-Provers on the Putnam Mathematical Competition." Neural Information Processing Systems (NeurIPS), 2024. [arXiv], [Code] |  |