I am a Ph.D. candidate in the School of Interactive Computing at Georgia Tech, advised by
Humphrey Shi.
I completed my Bachelor's in Computer Science and Engineering in 2023 at IIT Roorkee.
I am currently a High Performance AI Intern at NVIDIA (with Humphrey Shi), working on training agents for CUDA Kernel Optimization.
I have also interned at Allen AI (Summer 2025, with Christopher Clark), and Microsoft Research, Redmond (Summer 2024, with
Jianwei Yang).
My research trajectory so far can be broadly classified into three areas:
- I began my research career by developing models for dense pixel-level prediction tasks
[OneFormer,
SeMask].
- Building on this foundation, I worked on analyzing and improving the visual perception
capabilities of Multimodal Large Language Models
[VisPer-LM,
VCoder].
- Presently, my work focuses on developing agent systems for long-horizon tasks in real-world settings, leveraging reinforcement learning to have models learn effective multi-turn reasoning
[SAGE,
AUGUSTUS].
I'm seeking industry research roles starting in February 2027. My work focuses on agentic AI, multimodal reasoning, and RL infrastructure. If your team is hiring, I'd love to connect!
Reach out if you are interested in my research or would like to discuss any ideas. If you are a self-motivated researcher who's
looking for guidance on one of your projects or just want to chat about my experience, will you drop me an email?
Professional Life Happenings
- [July 2026]: Accepted to the ACM AI Leadership Summit Doctoral Consortium.
- [May 2026]: Passed my Ph.D. thesis proposal!
- [May 2026]: Invited Talk at Isola Lab@MIT! [Slides]
- [May 2026]: Accepted to the CVPR 2026 Doctoral Consortium.
- [February 2026]: SAGE is accepted to CVPR 2026! See you in Denver!
- [September 2025]: VisPer-LM (f.k.a. OLA-VLM) is accepted to NeurIPS 2025! See you in San Diego!
- [April 2025]: Passed my Ph.D. Qualifying Exam!