About

Hello

I’m a Member of Technical Staff at Skild AI, where I research and develop Vision-Language-Action (VLA) and World Action Models (WAM) for multi-embodiment robot control, toward generalist robot policies.

Before Skild, I was a Senior Research Scientist at Meta AI Research in Seattle, working on native image and video generation with large language models. Before that I was a postdoctoral researcher at Meta in New York with Ishan Misra, working on generation — Movie Gen, Emu Video, InstanceDiffusion — and evaluation (SelfEval).

I did my Ph.D. in Electrical Engineering at the University of Maryland, College Park with Rama Chellappa and Abhinav Shrivastava, on in-the-wild visual understanding: object discovery and localization, sparsely annotated detection, few-shot learning, tracking and re-identification. Before that, a dual degree in Electrical Engineering at IIT Kharagpur.

Service

• Conferences — CVPR ’19–’26, ECCV ’20–’26, ICCV ’19–’25, ICLR ’23–’25, NeurIPS ’23–’25, ICML ’24
• Journals — IJCV, T-PAMI, Pattern Recognition Letters, IEEE Access

Teaching

• Graduate TA, UMD — ENEE222 (Fall 2016), ENEE324 (Spring 2017), ENEE425 (Fall 2017), ENEE630 (Fall 2017)