At the Indian Institute of Technology Kharagpur, I implemented the core codebase for a cross-lingual NLP framework addressing content gaps between English and Hindi Wikipedia. The work used in-context learning with LLMs and machine translation.
I designed and executed automatic and human judgment-based evaluation pipelines, achieving up to 65% improvement in Hindi Wikipedia article quality. I also co-authored a peer-reviewed paper accepted at COLING 2025.
In an ongoing IIT Kharagpur collaboration, I’m contributing to research on bias removal in vision-language models. My research experience draws on an M.Sc. in Physics from Osmania University, where I graduated with distinction and a CGPA of 8.56.
I work with Python, statistical modeling, and data visualization, and have completed certifications in data analytics, machine learning, and deep learning.
