Léo R is a Senior Data Consultant based in Paris with 9 years of experience specializing in renewable energy markets, particularly French and UK Guarantees/REGOs and ROCs. He blends economics and machine learning to deliver actionable market forecasts and data-driven pricing insights, having modeled GO price dynamics using reservoir levels, weather, and trading history. His work spans policy-oriented research (OECD, French Ministry) and hands-on industry analysis at Veyt and Greenfact, enabling strategic recommendations for clients and stakeholders. As an active contributor to the skrub-data project, he improved a SimilarityEncoder for ML workflows—showing practical expertise in feature engineering and test-driven development. Beyond analytics, he engages in public service as a municipal counsellor in Plouzané, reflecting a commitment to local impact alongside technical proficiency. Fluent in bridging quantitative rigor with clear communication, he regularly presents findings to technical and non-technical audiences.
9 years of coding experience
Bachelor, Economics and Management, Mention très bien, Bachelor, Economics and Management, Mention très bien at University of Nice
Master, European Affairs, Economics and Public Policy, Master, European Affairs, Economics and Public Policy at Sciences Po
Master 1, European Studies, Mention bien, Master 1, European Studies, Mention bien at Sorbonne Nouvelle University
Two-year technical degree in International trade, International Trade, 15.47, Two-year technical degree in International trade, International Trade, 15.47 at ISIFA PLUS VALUES CFA
Master, Applied Economics, Master, Applied Economics at University of Paris
Technical Degree, Trade, Technical Degree, Trade at Ecofac Business School
Baccalaureate, Economics, Baccalaureate, Economics at Cornouaille High School
Contributions summary:Léo primarily contributed to the development and testing of the `SimilarityEncoder` class within the `skrub-data/skrub` repository. They added functionality for using the `HashingVectorizer` within the `SimilarityEncoder` and implemented tests for the added functionality. They refactored existing tests and added additional tests for different configurations, including tests for `n_prototypes`. Further, they refactored the code by subclassing `SimilarityEncoder` from `OneHotEncoder` and fixing a related import issue.
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.