Vishnu Penubarthi
Software Engineer at Meta
United States
Summary
🤩
Rockstar🎓
Top SchoolVishnu Penubarthi is a software engineer with 8 years of experience specializing in machine learning, model quantization, and autonomous vehicle research. He’s contributed to PyTorch core, adding model-reporting features and detectors that guide quantization decisions—a notable open-source impact on a high-profile ML framework. Vishnu has interned and worked across research and industry at MIT CSAIL, Meta, AWS, and Applied Intuition, building diagnostic APIs, visualization frameworks, and production improvements that measurably boosted system coverage and correctness. His academic background (BS and MEng from MIT) underpins published work on explaining multimodal errors in self-driving systems, and he’s taught machine learning to high-school students through MIT programs. Practical, research-driven, and product-minded, he blends low-level ML systems engineering with a track record of shipping tools that help practitioners debug and harden models.
8 years of coding experience
2 years of employment as a software developer
Master of Engineering - MEng Computer Science, Master of Engineering - MEng Computer Science at Massachusetts Institute of TechnologyShrewsbury High SchoolMassachusetts Academy of Math and Science
Telugu, English
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (8)
quantization10
pytorch10
machine-learning10
deep-learning10
python10
test-automation9
linear-algebra8
data-analysis7
Programming languages (4)
Github contributions (5)
pytorch/pytorch
Jun 2022 - Aug 2022
Tensors and Dynamic neural networks in Python with strong GPU acceleration
Role in this project:
Back-end Developer
Back-end DeveloperContributions:89 reviews, 105 commits, 55 PRs in 2 months
Contributions summary:Vishnu implemented features and tests for model report functionality within the PyTorch repository. They added a per-channel detector to the model report functionality to aid in the selection of dynamic vs. static quantization for linear layers. The user also added an outlier detector and the functionality for the ModelReport API to generate the Qconfig based on user suggestions. They also implemented modifications to support conv layers and their groups.
pythongpu-accelerationdeep-learninggpunumpy
vspenubarthi/EmerSave-R2
May 2018 - Jul 2020
Contributions:4 pushes, 1 branch in 2 years 2 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.