Academic

My current interests lie in AI Infrastructure and ML Systems, particularly large language model inference, inference optimization, GPU systems, and efficient model execution.

Previously, I worked on Lip-to-Speech research involving visual speech recognition, ASR-guided generation, diffusion models, and experimental analysis. This experience gave me hands-on experience in building experimental pipelines, evaluating model behavior, and analyzing why different methods succeed or fail.

I am especially interested in research that connects model behavior with system efficiency and reliability, and I am open to undergraduate research, research internships, and collaboration opportunities in related areas.