I’m pursuing a Master of Science in Robotics at Carnegie Mellon University’s Robotics Institute, advised by Dr. Jiaoyang Li. I work on skill-based, multi-agent, long-horizon task planning for robot manipulation, as part of a project with the Toyota Research Institute.
Before CMU, I spent almost two years as a founding engineer at a stealth robotics startup in San Francisco. I built a shared autonomy system for dual-arm manipulation that blends a teleoperator’s input with VLA policies, along with a runtime safety layer that hands control smoothly between the policy and the operator. I also designed a sensorized gripper for close-contact tasks with people, and built much of the teleoperation and data infrastructure behind it all: leader-follower arm systems, Unitree G1 and YAM integration, MuJoCo simulation, and Meta Quest and Apple Vision Pro control.
I graduated magna cum laude from New York University in 2025 with a double major in Computer Science and Data Science. My primary lab there was the Energy-Aware, Secure and Reliable Computing Research Group (EnSuRe), headed by Dr. Siddharth Garg, where I studied backdoor attacks on language models (an ICDE 2023 workshop paper) and built ICMLM, a language model for Indian classical music. I also worked with Dr. Carlos Fernandez-Granda at the Center for Data Science, training vision transformers on 3D MRI scans to predict neuropsychiatric symptoms of Alzheimer’s, and with Dr. Meredith Dank at the Marron Institute, analyzing data on runaway and homeless youth to help disrupt human trafficking. Outside the lab, I led NYU’s RoboMaster robotics competition team.
Beyond NYU, I worked with Dr. Aparna Varde at Montclair State University on DISCERN, a training-free way for robots to use commonsense knowledge to decide the order of tasks in human-robot collaboration, published at MIT URTC 2024 and IEEE Big Data 2024. Earlier, I worked with Dr. Shantanu Sharma at NJIT on information-theoretically secure search over encrypted databases (VLDB 2023), and with Dr. Sharad Mehrotra at UC Irvine on OccuV, a visualizer for spatial and temporal data, as an NSF REU intern.
I’ve also interned as a data scientist at INVIDI Technologies, forecasting ad impressions with RNNs, and built Kathak Saangi, a companion app for Kathak dancers with over 13,000 downloads.
My CV is available here.
Contact Information
- CMU Email
sroychou@andrew.cmu.edu- NYU Email
swag@nyu.edu
Other Links
- GitHub
@GOTWIC- Google Scholar
- Swagnik Roychoudhury
in/swagnik@swagnikr- Discord
@GOTWIC