David Bau
Machine Learning Engineer at Northeastern University
Boston, Massachusetts, United States
Summary
🤩
RockstarDavid Bau is a Machine Learning Engineer and Northeastern professor with 19 years of experience specializing in interpretable deep networks, NLP, and vision, and an ex-Google background. Based in Boston, he combines academic rigor with practical engineering—contributing production-ready PyTorch tools for semantic segmentation and GAN neuron visualization (gandissect) and maintaining widely used utilities like a seeded JavaScript RNG. He advocates for transparency in deep learning and bridges research and deployment through full-stack contributions that improve reproducibility and sharing (e.g., Colab setups and Twitter card support). His work reflects a rare blend of theory, open-source impact, and attention to usability across both model interpretability and developer tooling.
19 years of coding experience
Github Skills (28)
webbrowser10
algorithm10
algorithms10
pytorch10
javascript10
python10
dcgan10
semantic-segmentation10
machine-learning10
cgan10
browser10
randomization10
computer-vision10
jupyter-notebook10
web-development10
Programming languages (9)
Github contributions (5)
davidbau/seedrandom
May 2013 - Sep 2019
seeded random number generator for Javascript
Role in this project:
Backend Developer
Backend DeveloperContributions:145 commits, 16 PRs, 73 pushes in 6 years 5 months
Contributions summary:David primarily focused on the development of the `seedrandom.js` library, a seeded random number generator for JavaScript. The contributions include implementing and refactoring the core algorithm, adding features, and fixing bugs. The user also worked on creating and updating the documentation to reflect the changes and new capabilities of the library. The user added support for node.js crypto and browser environments, which is crucial for general utility.
javascript
CSAILVision/gandissect
Nov 2018 - Aug 2020
Pytorch-based tools for visualizing and understanding the neurons of a GAN. https://gandissect.csail.mit.edu/
Role in this project:Full-stack Developer
Full-stack DeveloperContributions:141 commits, 3 PRs, 44 pushes in 1 year 8 months
Contributions summary:David contributed significantly to the project by adding features related to Twitter card support and public host configuration, impacting the project's visibility and sharing capabilities. They modified the server-side code to integrate these functionalities, including updates to server-side logic and file handling. Furthermore, the user enhanced the project by fixing support for built-in Python functions, ensuring broader compatibility and stability. The user's work suggests a focus on both the front-end presentation and back-end functionality, aligning with full-stack development.
pytorchganimage-manipulationdeep-learninginteractive-visualizations
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.