Ilya Mironov
Senior Staff Research Scientist at Meta
San Francisco, California, United States
Summary
🤩
Rockstar🎓
Top SchoolIlya Mironov is a Senior Staff Research Scientist in San Francisco with 11+ years of experience applying rigorous research to practical ML and privacy engineering at Meta, Google, and Microsoft. He holds a PhD in computer science from Stanford and a long-standing background in theoretical and applied research dating back to the 1990s. Ilya blends hands-on implementation and code hygiene—evidenced by contributions to high-profile open-source projects like TensorFlow Privacy and Google’s RAPPOR—with leadership experience managing research and engineering teams. His work focuses on differential privacy and privacy-preserving ML, where he has improved tutorials, refactored core models, and optimized simulation tools that enable real-world adoption. Colleagues benefit from his rare mix of deep academic training and pragmatic software craftsmanship that ships robust, auditable systems.
11 years of coding experience
21 years of employment as a software developer
M.S., computer science, M.S., computer science at Saint Petersburg State UniversityPh.D., computer science, Ph.D., computer science at StanfordMath, Physics, Math, Physics at Saint Petersburg Lyceum 239
French, Russian, English
Github Skills (17)
python10
model-driven10
r10
machine-learning10
model-building10
privacy10
differential-privacy10
tensorflow10
modeling10
model-driven-development10
data-analysis10
data-science9
statistical-models9
algorithms8
algorithm4
Programming languages (7)
Github contributions (5)
google/rappor
Mar 2015 - Jun 2016
RAPPOR: Privacy-Preserving Reporting Algorithms
Role in this project:
Back-end Developer & Data Scientist
Back-end Developer & Data ScientistContributions:103 commits, 22 PRs, 71 pushes in 1 year 3 months
Contributions summary:Ilya primarily contributed to the `tests/gen_counts.R` file, which includes code for simulating RAPPOR reports. They added new functions for fast simulation of reports and optimized existing code. The user also refactored the code for improved clarity and maintainability. This work aligns with the repository's purpose of privacy-preserving reporting algorithms.
securityprivacy-preservingreportingprivacy
tensorflow/privacy
Jan 2019 - Jun 2019
Library for training machine learning models with privacy for training data
Role in this project:ML Engineer
ML EngineerContributions:10 commits, 11 comments, 9 issues in 5 months
Contributions summary:Ilya contributed to the TensorFlow Privacy library, making several updates and improvements. Their work included modifying the tutorial for differentially private training with MNIST, fixing import issues, and restoring the ability to run a script as a standalone tool. Additionally, the user updated the privacy analysis in the logistic regression tutorial and cleaned up code to replace deprecated endpoints. These changes demonstrate a focus on improving the usability and functionality of the library for differentially private machine learning.
machine-learning-trainingprivacydifferential-privacymachine-learningtraining
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.