Hi, I'm Kabir J.

Github Linkedin Twitter Goodreads

Data Science Student and Student Researcher.

My broader interests lie in interpretability, alignment, SSL techniques, pretraining in LLMs and MLLMs. I’m currently working on interpretability in Multimodal Large Language Models (MLLMs) and am curious about how models think, reason, fail, and can be improved.

Right now, I'm:

Reach out to work on projects and collaborate on ML research — email me

Publications

Projects

Other/Fun

Res