I am a Research Fellow at Sarvam AI, working on foundation models.

My past research focuses on efficient pre-training and evaluation of graph–language multimodal models, alongside exploring interpretability and alignment in real-world scenarios using the linear representation hypothesis.

Topics that intrigue me include (but aren’t limited to): mechanistic interpretability, alignment, graph-language modeling, deterministic & stochastic reasoning mechanisms, or anything that imparts tacit knowledge to models!

Outside academics and research, I love cinema and music: critically analyzing modern films, filmography, and directing. My team and I have won competitions at BITS-H, Ashoka, and internal university events. I’m also passionate about both sports and e‑sports, having competed and worked as an analyst.

I also blog (on LessWrong) and tweet (on X) sometimes.