|

Hi, I'm Jannat, a PhD researcher in Computer Science at VCU. I work on vision models and vision-language models, mostly around making visual and scientific information more accessible.

Most of what I do comes down to building efficient, low-compute multimodal models, with blind and low-vision users in mind from the start rather than as an afterthought. I co-authored a paper at BioASQ 2025 (CLEF) on biomedical information extraction.

Before this I did a bit of everything, machine learning models, multimodal AI accessibility, efficient models, RAG pipelines, full-stack apps, some NLP on the side.

Off the clock: I believe learning new skills keeps my brain active, so I'm always picking up something new, though tennis is my all-time favorite, where the ball make my body move.