Elie Bursztein
Founder at google
Mountain View, California, United States
Summary
🤩
RockstarElie Bursztein is a founder and AI cybersecurity expert with 11 years of engineering experience based in Mountain View, CA, focused on making AI safe and secure through work at Google and DeepMind. He blends deep ML engineering—evidenced by contributions to high-profile open-source projects like TensorFlow Similarity and Keras Tuner—with pragmatic backend development and performance optimizations such as multi-GPU support and reimplemented loss functions. Elie’s work spans research-grade model tooling and production-ready systems, enabling faster similarity learning and scalable hyperparameter search. As an entrepreneur and founder of the Etteilla Art Foundation, he brings a creative, interdisciplinary perspective to technical problems. He is particularly adept at refactoring complex codebases to improve maintainability and training efficiency, a skill that surfaces across his open-source contributions.
11 years of coding experience
Github Skills (10)
metric-learning10
keras10
hyperparameter-optimization10
machine-learning10
deep-learning10
tensorflow10
automl10
indexer10
python10
cosine-similarity9
Programming languages (7)
Github contributions (5)
keras-team/keras-tuner
Jan 2018 - Sep 2019
A Hyperparameter Tuning Library for Keras
Role in this project:
Back-end Developer & ML Engineer
Back-end Developer & ML EngineerContributions:146 commits, 6 PRs, 3 pushes in 1 year 8 months
Contributions summary:Elie reworked a version of the kerastuner library, implementing a new API for hyperparameter tuning. They focused on implementing key functionalities related to the engine, including the `HyperTuner` meta-class, instance, and execution. The user demonstrated expertise in integrating the model with the provided search algorithms, including the RandomSearch. They also implemented multi-GPU support to enhance performance.
hyperparameterdeep-learningmachine-learningautomltensorflow
tensorflow/similarity
Oct 2020 - May 2022
TensorFlow Similarity is a python package focused on making similarity learning quick and easy.
Role in this project:Back-end Developer & ML Engineer
Back-end Developer & ML EngineerContributions:5 releases, 11 reviews, 133 commits in 1 year 7 months
Contributions summary:Elie primarily focused on code cleanup and refactoring within the `tensorflow/similarity` repository, specifically within the `indexer` and `losses` modules. These changes involved modifying files related to indexing functionality and loss calculations. They also reimplemented a triplet loss function, indicating an effort to optimize the training process.
pythonsimilarity-searchmetric-learningtensorflowbarlow-twins
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.