Isht Dwivedi
Engineering at Applied Intuition
San Jose, California, United States
Summary
👤
Senior🎓
Top SchoolIsht Dwivedi is a computer vision engineer with nine years of experience focused on perception and scene understanding for autonomous vehicles, currently working at Applied Intuition in San Jose. He has led camera-only HD map creation and road-condition monitoring efforts at Honda Research Institute, and earlier built research-grade vision and trajectory-prediction systems addressing unstructured scenarios like construction zones. With a Columbia M.S. in Computer Science and a strong academic publication record (including an ACM Multimedia paper), he blends rigorous research with production-minded engineering. He also contributes to open-source ML tooling—converting a Caffe DeepLab-ResNet into PyTorch to enable flexible pre-trained model reuse—showing a knack for practical model adaptation and data scaling. Notably, his background spans sketch segmentation, particle-physics reconstruction, and multilingual input tools, reflecting a rare breadth across vision, algorithms, and systems.
9 years of coding experience
8 years of employment as a software developer
Bachelor of Technology (B.Tech.) Major in Electrical and Electronics Engineering Minor in Computer Science and Engineering, Bachelor of Technology (B.Tech.) Major in Electrical and Electronics Engineering Minor in Computer Science and Engineering at Indian Institute of Technology, GuwahatiMaster's degree Computer Science, Master's degree Computer Science at Columbia University
English, Hindi
Stackoverflow
65reputation
12kreached
1answer
6questions
Github Skills (9)
semantic-segmentation10
computer-vision10
pytorch10
caffe10
deep-learning10
python10
android6
androidplot6
android-layout6
Programming languages (6)
Github contributions (5)
isht7/pytorch-deeplab-resnet
Apr 2017 - Nov 2021
DeepLab resnet v2 model in pytorch
Role in this project:
ML Engineer
ML EngineerContributions:107 commits, 2 PRs, 100 pushes in 4 years 7 months
Contributions summary:Isht primarily contributed to converting and integrating a pre-trained Caffe-based DeepLab-ResNet model into a PyTorch framework. They wrote a script to convert Caffe model parameters to PyTorch, enabling the use of the model within the PyTorch ecosystem. Furthermore, they modified the model to support custom data and image scaling, suggesting a focus on adapting the model for flexible use cases and data inputs.
pascal-vocpytorchsemantic-segmentationdeeplabdeep-learning
isht7/isht7.github.io
Jan 2021 - Mar 2025
Contributions:52 pushes in 4 years 3 months
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.