Summary
Harshit Gupta is a software engineer at Google's Augmented Reality team with 8 years of experience bridging deep learning, computational imaging, and 3D computer vision to solve real-world AR problems. He holds a PhD from EPFL where he developed supervised and unsupervised methods for 2D, 3D and 3D+time reconstruction, including physics-aware GANs and trainable-activation networks, and continued that line of work as a postdoc at SLAC building autograd-enabled physics models for dynamic 3D reconstruction. Comfortable moving ideas from research to production, he combines strong theoretical foundations with practical Python/PyTorch implementation experience. Based in Sunnyvale, he blends academic rigor with industry impact, uniquely pairing inverse-problem expertise with hands-on AR system development.
8 years of coding experience