Roman B
Paris, Ile-de-France
Summary
🤩
Rockstar🎓
Top SchoolRoman B is a Machine Learning Engineer based in Paris with six years of experience building privacy-preserving ML systems. At Zama he works on practical FHE deployments and contributed Poisson Regression support to the open-source Concrete-ML framework, including tutorials that bridge Scikit-Learn GLMs with fully homomorphic execution. Trained as an engineer in computer science and applied mathematics (Grenoble, MSIAM, and an Erasmus at Instituto Superior Técnico), he blends strong mathematical foundations with hands-on software delivery. Roman focuses on making advanced cryptographic ML techniques accessible to practitioners, turning research-grade tools into usable libraries and documentation. Comfortable across the stack from core library changes to end-user examples, he has a track record of shipping reproducible, well-documented features in open-source ML tooling.
6 years of coding experience
Engineering degree, Computer Science and Applied Mathematics, Engineering degree, Computer Science and Applied Mathematics at National School of Computer Science and Applied Mathematics of GrenobleScience Track French Baccalauréat | American High School Diploma, Sciences, Science Track French Baccalauréat | American High School Diploma, Sciences at Lycée Français de San FranciscoInternational Master of Science in Industrial and Applied Mathematics (MSIAM), Applied Mathematics and Computer Science, International Master of Science in Industrial and Applied Mathematics (MSIAM), Applied Mathematics and Computer Science at Université Grenoble AlpesIntensive two years preparing for the national competitive exam to join Engineering Schools, Maths, Physics, Engineering Sciences and Computer Science, Intensive two years preparing for the national competitive exam to join Engineering Schools, Maths, Physics, Engineering Sciences and Computer Science at Lycée ChampollionErasmus exchange, Erasmus exchange at Instituto Superior Técnico
French, English, Spanish
Github Skills (8)
scikit10
machine-learning10
python10
data-science10
scikit-learn10
documentation9
pytorch8
tensorflow8
Programming languages (4)
Github contributions (5)
zama-ai/concrete-ml
Apr 2022 - Jan 2023
Concrete ML: Privacy Preserving ML framework using Fully Homomorphic Encryption (FHE), built on top of Concrete, with bindings to traditional ML frameworks.
Role in this project:
ML Engineer
ML EngineerContributions:1530 reviews, 118 commits, 531 PRs in 9 months
Contributions summary:Roman implemented and integrated Poisson Regression models into the `concrete-ml` library. They added the Poisson Regression implementation to the library and created a tutorial to showcase the training of Generalized Linear Models (GLM) with a Poisson distribution within the Concrete-ML framework, including running the model in FHE, and integrating Scikit-Learn's PoissonRegressor. This involved code modifications within the core library, and the creation of documentation and advanced examples for users.
scientistspythonparticulartfhefhe
zama-ai/blog
Jul 2023 - Aug 2023
Public repo for HF blog posts
Contributions:11 pushes, 1 branch in 16 days
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.