Madhav Kanda
Hello! I am a fully funded M.S. Computer Science (Thesis Track) student at the University of Illinois Urbana-Champaign. My research interests lie in multimodal multi-agent systems, error-free LLM-driven code generation, and program repair using LLMs.
I am currently working with Prof. Sasa Misailovic on enhancing the semantic accuracy of LLM-generated code through grammar-guided generation. Additionally, as part of my CS598 project, I am working under the guidance of Prof. Lingming Zhang on automatic software debugging using small language models. Previously, I collaborated with Prof. Heng Ji and Prof. Unnat Jain to design a multi-agent framework for complex VQA tasks—this work was recently submitted to ACL ‘25!
I earned my BTech with Honours in Computer Science and Engineering from IIT Gandhinagar, where I conducted research under the guidance of Prof. Nipun Batra and Prof. Shanmuganathan Raman on probabilistic machine learning (and its applications) and generative adversarial networks, respectively.
During my second-year summer internship, I had the opportunity to work with Prof. A.G. Ramakrishnan at IISc Bangalore, focusing on natural language processing tasks. In my third-year summer internship, I worked with Prof. Aki Vehtari at Aalto University on reducing divergences encountered during Hamiltonian Monte Carlo sampling. I also worked as a computer vision intern at Spyne.ai, where I focused on Virtual Try-On. In addition to my research, I was featured as a contributor in Dr. Kevin Murphy’s book, Probabilistic Machine Learning: Advanced Topics, in recognition of my contributions.
I’m actively seeking internship opportunities for Summer 2025. If you believe I’d be a great fit for your team, feel free to reach out at madhav3@illinois.edu. I’d love to connect!
Affiliations
IIT Gandhinagar
2020-2024
Spyne.ai
2022
Indian Institute of Science
2022
Aalto University
2023
UIUC
2024-2026