Arpit Bansal
College Park, Maryland, United States
Summary
🤩
RockstarArpit Bansal is a research scientist with a PhD in Computer Science from the University of Maryland and nine years of industry and academic experience focused on image synthesis and algorithmic neural network design. He has published at NeurIPS, ICLR, ICML, and CVPR and has applied his research in internships and roles at Meta (vision encoder for image reasoning and high-precision editing), AWS AI Labs (text-to-image synthesis), and now Google. His work spans diffusion and autoregressive models for text-to-image generation, bridging cutting-edge research with production-minded implementation. Trained at IIT Kharagpur and UMD, he pairs strong theoretical grounding (3.925 GPA in his PhD) with practical engineering experience from enterprise roles at Visa. An underappreciated thread in his profile is repeated success translating novel research into applied systems across both cloud and product-focused teams.
9 years of coding experience
Github Skills (9)
pytorch7
transfer-learning6
jekyll-theme3
diffusion-probabilistic1
machine-learning1
multimodal1
image-classification1
denoising1
neural-network1
Programming languages (2)
Github contributions (5)
arpitbansal297/Cold-Diffusion-Models
Aug 2022 - Oct 2022
Official implementation of Cold-Diffusion for different transformations in pytorch.
Contributions:16 commits, 1 PR, 4 pushes in 1 month
pytorch
arpitbansal297/Universal-Guided-Diffusion
Jan 2023 - Mar 2023
Contributions:16 commits, 3 pushes, 2 branches in 1 month
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.