University of California, San DiegoI am a second-year graduate student at UC San Diego working with Loris D'Antoni. My current research focuses on constrained generation for language models and their applications to compiler testing.
Previously, I worked on low-latency trading systems at J.P. Morgan and Morgan Stanley.
Education
-
University of California, San DiegoM.S. in Computer Science
Advisor: Loris D'AntoniSep. 2024 - Jun. 2026 (Expected) -
P.S.G. College of TechnologyB.E. in Computer Science and Engineering
Advisor: G. R. KarpagamJun. 2019 - May 2023
Experience
-
Morgan StanleyAssociateJul. 2023 - Aug. 2024 -
Morgan StanleySoftware Engineer InternJan. 2023 - Jun. 2023
-
J.P. Morgan Chase & Co.Software Engineer InternJun. 2022 - Jul. 2022
News
Selected Publications (view all )
Bootstrapping Fuzzers for Compilers of Low-Resource Language Dialects Using Language Models
Sairam Vaidya, Marcel Böhme, Loris D'Antoni
arXiv Preprint 2025
We present Germinator, a dialect-agnostic and dialect-effective fuzzing approach for extensible compilers like MLIR. By automatically extracting grammars from dialect specifications and using LLMs to generate diverse seed inputs, Germinator bootstraps coverage-guided fuzzing without manual effort. Evaluated on six MLIR projects spanning 91 dialects, it improved line coverage by 10-120% and discovered 88 previously unknown bugs.
Bootstrapping Fuzzers for Compilers of Low-Resource Language Dialects Using Language Models
Sairam Vaidya, Marcel Böhme, Loris D'Antoni
arXiv Preprint 2025
We present Germinator, a dialect-agnostic and dialect-effective fuzzing approach for extensible compilers like MLIR. By automatically extracting grammars from dialect specifications and using LLMs to generate diverse seed inputs, Germinator bootstraps coverage-guided fuzzing without manual effort. Evaluated on six MLIR projects spanning 91 dialects, it improved line coverage by 10-120% and discovered 88 previously unknown bugs.
Constrained Sampling for Language Models Should Be Easy: An MCMC Perspective
Emmanuel Anaya Gonzalez*, Sairam Vaidya*, Kanghee Park, Ruyi Ji, Taylor Berg-Kirkpatrick, Loris D'Antoni (* equal contribution)
NeurIPS 2025
We propose a new constrained sampling framework based on Markov Chain Monte Carlo (MCMC) that is constraint satisfying, monotonically converging to the true conditional distribution, and efficient at generating high-quality samples in few steps.
Constrained Sampling for Language Models Should Be Easy: An MCMC Perspective
Emmanuel Anaya Gonzalez*, Sairam Vaidya*, Kanghee Park, Ruyi Ji, Taylor Berg-Kirkpatrick, Loris D'Antoni (* equal contribution)
NeurIPS 2025
We propose a new constrained sampling framework based on Markov Chain Monte Carlo (MCMC) that is constraint satisfying, monotonically converging to the true conditional distribution, and efficient at generating high-quality samples in few steps.