Pavithra Vijayakrishnan
Senior Software Engineer at Google
Sunnyvale, California, United States
Summary
🤩
Rockstar🎓
Top SchoolPavithra Vijayakrishnan is a Senior Software Engineer with a decade of experience building robust backend systems and ML infrastructure, currently working on Google Discover Ads in Sunnyvale. She combines strong academic foundations—a B.E. in Computer Science with top honors and an M.S. from University of Wisconsin–Madison—with hands-on engineering at scale. Pavithra is an active open-source contributor to high-profile ML projects like Keras and TensorFlow, where her work on loss functions, estimator loss reduction, and documentation improved correctness and usability for the broader community. Her contributions show a focus on stabilizing core training components and clarifying complex examples, bridging engineering and technical writing. Known for incremental yet impactful fixes, she brings a pragmatic approach to improving long-lived libraries and production systems. Colleagues rely on her for careful, standards-aligned refactors that reduce technical debt while preserving model semantics.
10 years of coding experience
Master of Science (M.S.), Computer Science, 3.6/4.0, Master of Science (M.S.), Computer Science, 3.6/4.0 at University of Wisconsin-MadisonAnna University, Chennai
Tamil, English
Github Skills (15)
neural-network10
keras10
machine-learning10
deep-learning10
tensorflow10
python10
documentation10
basics9
learn-ruby-on-rails9
user-manual9
unit-testing9
pytorch9
api8
apidoc8
deep-neural-networks6
Programming languages (3)
Github contributions (5)
tensorflow/estimator
Nov 2018 - Apr 2020
TensorFlow Estimator
Role in this project:
Back-end Developer & ML Engineer
Back-end Developer & ML EngineerContributions:14 commits in 1 year 5 months
Contributions summary:Pavithra primarily contributed to the `tensorflow/estimator` repository by updating and refactoring code related to loss reduction in various estimator implementations. Their changes focused on aligning the loss reduction behavior with TensorFlow V2 and involved modifications across multiple files within the estimator framework. They also updated the loss calculation logic in test files, showcasing a deep understanding of the underlying estimator functionalities. Moreover, the user was involved in code cleanups, removing usages of V1 metrics and updating the AUC enums inputs.
deep-learningmachine-learningtensorflowtensorflow-estimatorestimator
keras-team/keras
Jan 2018 - Sep 2019
Deep Learning for humans
Role in this project:Back-end Developer
Back-end DeveloperContributions:36 commits, 17 PRs, 3 pushes in 1 year 7 months
Contributions summary:Pavithra primarily contributed to bug fixes and code improvements within the Keras library. Their work involved addressing minor issues in example code, refining convolutional layer implementations, and correcting documentation errors in the training engine. They also made significant additions by adding Loss, LossFunctionWrapper, and MeanSquaredError classes, expanding the available loss functions within the library. The user's commits demonstrate a focus on improving the functionality, documentation, and stability of the core Keras codebase.
pythondata-sciencedeep-learningneural-networksmachine-learning
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.