CARVIEW

MOTORHOMES

Select Language

HTTP/2 200 server: GitHub.com content-type: text/html; charset=utf-8 last-modified: Sat, 04 Oct 2025 04:12:52 GMT access-control-allow-origin: * strict-transport-security: max-age=31556952 etag: W/"68e09ec4-6516" expires: Wed, 31 Dec 2025 02:58:00 GMT cache-control: max-age=600 content-encoding: gzip x-proxy-cache: MISS x-github-request-id: F448:234FE9:ABAB79:C0FB15:69548EE0 accept-ranges: bytes age: 0 date: Wed, 31 Dec 2025 02:48:00 GMT via: 1.1 varnish x-served-by: cache-bom-vanm7210044-BOM x-cache: MISS x-cache-hits: 0 x-timer: S1767149280.425470,VS0,VE205 vary: Accept-Encoding x-fastly-request-id: b09db9e0aefede7c12b5927e9820866f744151f8 content-length: 5670 Ryan Punamiya

Ryan Punamiya

B.S. Computer Science

Georgia Institute of Technology

College of Computing

rpunamiya6 [AT] gatech.edu

About

Hi I'm Ryan! I'm a fourth-year B.S. CS major at Georgia Tech. I am a researcher at the RL2 Lab, advised by Prof. Danfei Xu and the Hoffman Lab, advised by Prof. Judy Hoffman. I am also fortunate to be mentored by Simar Kareer, PhD at Georgia Tech. I am currently interested in studying scalable approaches for task-to-task transfer and generalization in robot learning. Most recently, I have been working on enabling different types of transfer from egocentric human data.

I am currently looking for PhD positions for Fall 2026.

News

Sept 2025 Our work, EgoBridge was accepted to NeurIPS 2025 and was nominated for an Oral presentation at the H2R Workshop at CoRL 2025.

Sept 2025 Received the President's Undergraduate Research Award (PURA) for my research in robotics.

Aug 2025 We publicly released EMMA, our recent endeavour in scaling mobile manipulation through egocentric human video.

Feb 2025 Our work, EgoMimic, was showcased in a Meta AI blog post highlighting its role in embodied AI with Project Aria and Ego4D.

Oct 2024 Our paper, EgoMimic was showcased at the LFDM and X-Embodiment Workshops at CoRL 2024.

Sep 2024 Received the President's Undergraduate Research Award for my research in robotics.

Education

Georgia Institute of Technology Atlanta, GA, USA

B.S in Computer Science Aug. 2022 - Present

GPA: 4.0/4.0

Advisors: Danfei Xu, Judy Hoffman

Publications

( indicates equal contribution )*

2025

EgoBridge: Domain Adaptation for Generalizable Imitation from Egocentric Human Data

Ryan Punamiya, Dhruv Patel, Patcharapong Aphiwetsa, Pranav Kuppili, Lawrence Y. Zhu, Simar Kareer*, Judy Hoffman*, Danfei Xu*

NeurIPS 2025, CoRL 2025 H2R Workshop (Oral)

Paper Website

EMMA: Scaling Mobile Manipulation via Egocentric Human Data

Lawrence Y. Zhu, Pranav Kuppili*, Ryan Punamiya*, Patcharapong Aphiwetsa, Dhruv Patel, Simar Kareer, Sehoon Ha, Danfei Xu

Preprint

Paper Website

2024

EgoMimic: Scaling Imitation Learning via Egocentric Video

Simar Kareer, Ryan Punamiya*, Dhruv Patel*, Pranay Mathur*, Shuo Cheng, Chen Wang, Judy Hoffman, Danfei Xu

ICRA 2025

Paper Website Code

Projects

Partnr - Intelligent co-pilot for consultants

Overall winner of AI ATL 2023 Hackathon

Website

RoadSense - Embedded system for passive pothole detection in cars

Winner of Georgia Tech Grand Challenges Grant ($700)

Website

Personal

Outside of robotics and classes you can find me:

Climbing (Bouldering)
at the Gym (I love weight training to destress)
making coffee and latte art
Reading
clicking photos

Vitæ

Full CV in PDF.

Hoffman Lab Aug 2024 - Present

Student Researcher (Advisor: Judy Hoffman)
Domain Adaptation for cross embodiment robot learning
RL2 Lab Aug 2023 - Present

Student Researcher (Advisor: Danfei Xu)
Cross embodiment learning for manipulation in robotics
LIDAR Lab Jan 2023 - May 2024

Student Researcher - Georgia Tech VIP Program (Advisor: Ye Zhao)
Generative models for quadruped locomotion
Clyd.ai Oct 2023 - Apr 2024

Co-founder and CTO
AI powered dashboard for consultants
NLP X Lab Jan 2023 - May 2023

Student Assistant (Advisor: Wei Xu)
Multi-lingual natural language processing with a focus on Hindi
Georgia Institute of Technology Aug 2022 - Present

B.S. in Computer Science

Original Source | Taken Source