Olivier Grisel

Also known as ogrisel

Olivier Grisel is a machine learning engineer who has worked on scikit-learn since 2010, a Python module used by machine learning teams on tabular data. He became a scikit-learn core developer in 2010 and earned an MSc in Advanced Computing from Imperial College London and an MSc in Mathematics and Computer Science from ENSTA ParisTech. At Nuxeo, he worked on text analytics and natural language processing, and he coauthored scikit-learn papers in 2011 and 2013. He worked full-time on scikit-learn at Inria, presented histogram-based gradient boosting in 2019, joined Probabl in 2024, and coauthored a 2025 SurvivalBoost paper. As of October 2026, he is a Machine Learning Engineer in Probabl's Open Source team and a member of scikit-learn's maintainers team.

Now at: Probabl
Works: 54
Citations: 73,408
h-index: 20
Badges: Rising
Bridge
Links: GitHub

Publication numbers from OpenAlex, as of 22 October 2026.

Timeline

Since 2010

A scikit-learn core developer

He becomes a scikit-learn core developer in 2010. He earns an MSc in Advanced Computing from Imperial College London and an MSc in Mathematics and Computer Science from ENSTA ParisTech.
2011

R&D engineer at Nuxeo

In 2011, he is an R&D engineer at Nuxeo, working on text analytics, natural language processing and semantic technologies for document management. He is active in AFPy, the French-speaking Python user group, and contributes to scikit-learn.
2011–2013

Two scikit-learn papers

In 2011, he coauthors “Scikit-learn: Machine Learning in Python” with Fabian Pedregosa, Gaël Varoquaux and others; the paper appears in JMLR 12. In 2013, he coauthors “API design for machine learning software: experiences from the scikit-learn project” with Lars Buitinck, Gilles Louppe and others.
2013–2017

Full-time on scikit-learn at Inria

Inria funds him from 2013 to 2017 to work full-time on scikit-learn. In 2016, he is a software engineer at Inria Parietal, improving scikit-learn and related tools.
2019

Gradient boosting in scikit-learn 0.21

At EuroPython 2019 in Basel, he presents the histogram-based Gradient Boosted Trees in scikit-learn 0.21. The implementation competes with XGBoost and LightGBM on multi-core CPUs; the talk introduces pygbm, a numba-based prototype for the scikit-learn implementation. He is a machine learning engineer at Inria Saclay, near Paris.
2023–2024

Inria spins off Probabl

In 2023, Inria spins off Probabl as a mission-driven company to take scikit-learn beyond a research lab. In February 2024, he is among the researchers joining Probabl. Scikit-learn is a Python module used by machine learning teams on tabular data.
2025

SurvivalBoost for competing risks

In 2025, he coauthors “Survival Models: Proper Scoring Rule and Stochastic Optimization with Competing Risks.” The paper presents SurvivalBoost, a gradient-boosted tree method for survival and competing-risks analysis.
As of October 2026

Maintaining scikit-learn at Probabl

He is a Machine Learning Engineer in Probabl's Open Source team and a scikit-learn contributor. He is also part of the scikit-learn maintainers team, which consolidates the library's development and maintenance.