Agrim Gupta
Research Scientist at Meta
San Francisco Bay Area United States
Summary
🤩
Rockstar🎓
Top SchoolAgrim Gupta is a research scientist with a decade of experience building multimodal AI and robotics systems, currently at Meta’s TBD Labs after recent research roles at DeepMind and Google. He completed a PhD at Stanford focused on video generation, world models, and self-supervised representation learning, and has taught deep learning for computer vision (CS231n). Agrim has co-led high-impact projects like RoboCat and contributed to foundation models for robot learning (VIMA, MetaMorph), blending theory with robot manipulation and real-world deployment. His open-source work includes backend contributions to SNAP, adding cascade detection and graph traversal capabilities to a widely used network analysis library. Comfortable moving between core algorithm development and applied systems, he combines strong academic rigor with hands-on engineering across industry research labs. A less obvious strength is his sustained cross‑institutional collaboration—bridging academic, corporate, and open-source efforts to accelerate multimodal and robotic research.
10 years of coding experience
6 years of employment as a software developer
BITS Pilani, Birla Institute of Technology and ScienceDoctor of Philosophy - PhD Computer Science, Doctor of Philosophy - PhD Computer Science at Stanford University
Stackoverflow
11reputation
2kreached
0answers
1question
Github Skills (12)
algorithm10
data-structures10
algorithms10
graph-algorithms10
c-language10
cprogramming-language10
network-analysis10
data-structure10
binary-search9
keras6
recurrent-neural-networks6
tensorflow6
Programming languages (4)
Github contributions (5)
snap-stanford/snap
Apr 2016 - Jun 2016
Stanford Network Analysis Platform (SNAP) is a general purpose network analysis and graph mining library.
Role in this project:
Back-end Developer
Back-end DeveloperContributions:6 commits in 2 months
Contributions summary:Agrim primarily contributed to the development of network analysis and graph mining functionalities within the SNAP library. Their work involved implementing and modifying core functions related to cascade detection and graph algorithms, as evidenced by the addition of `CascGraph` and `CascFind` methods. The code changes include implementing binary search algorithms and modifying functions related to graph traversal.
stanfordpythonminingpurposegraph
agrimgupta92/sgan
Jun 2018 - Oct 2019
Code for "Social GAN: Socially Acceptable Trajectories with Generative Adversarial Networks", Gupta et al, CVPR 2018
Contributions:2 commits, 1 PR, 1 push in 1 year 4 months
pytorchhuman-trajectory-predictiondeep-learningadversarialgenerative-adversarial-network
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.