Antoine Collas
Models Team Lead at Karavela
Greater Paris Metropolitan Region France
Summary
👤
Senior🎓
Top SchoolAntoine Collas is a models team lead with nine years of experience at the intersection of machine learning, signal processing and neuroscience, currently building foundation brain‑decoding models at Karavela from next‑generation MRI. He holds a PhD in signal processing and applied math from CentraleSupélec and has a strong research record in Riemannian geometry, domain adaptation and generative modeling, with publications and a NeurIPS spotlight among his highlights. As a postdoc at Inria he translated geometric methods to MEG/EEG and fMRI applications and maintains open‑source benchmarking tools (SKADA, SKADA‑Bench) used by the community. Antoine also contributes to well‑known scientific libraries like geomstats and pymanopt, implementing statistical manifolds and rigorous unit tests, reflecting a rare blend of theoretical depth and production‑grade software engineering.
9 years of coding experience
2 years of employment as a software developer
Computer Science, Computer Science at UTSEUSDiplôme d'ingénieur, Génie informatique, Diplôme d'ingénieur, Génie informatique at Université de Technologie de CompiègneBaccalauréat scientifique spécialité maths, section européenne anglaise, Baccalauréat scientifique spécialité maths, section européenne anglaise at Lycée international Charles de Gaulle - DijonPhD, Signal processing / Computer science / Applied mathematics, PhD, Signal processing / Computer science / Applied mathematics at Université Paris-Saclay
French, English
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (15)
testing10
unit-testing10
statistics10
scipy10
probability-distribution10
manifold10
python10
distributions10
numpy10
geom10
optimization9
geometry9
machine-learning9
geo9
deep-learning8
Programming languages (5)
Github contributions (5)
pymanopt/pymanopt
Apr 2020 - May 2020
Python toolbox for optimization on Riemannian manifolds with support for automatic differentiation
Role in this project:
Back-end Developer & QA Engineer
Back-end Developer & QA EngineerContributions:1 release, 14 reviews, 7 commits in 1 month
Contributions summary:Antoine primarily contributed to the development and testing of mathematical manifolds within the `pymanopt` library. Their work included implementing new manifolds, specifically a complex Grassmann manifold and a manifold for strictly positive vectors. The user also wrote tests for these manifolds, ensuring their functionality and correctness. Additionally, the user made minor code corrections, such as fixing typos and correcting code formatting.
pythonautomatic-differentiationdifferentiationtoolboxoptimization
geomstats/geomstats
Oct 2022 - Oct 2022
Computations and statistics on manifolds with geometric structures.
Role in this project:Data Scientist
Data ScientistContributions:2 reviews, 16 commits, 1 PR in 9 days
Contributions summary:Antoine contributed significantly to the implementation of statistical manifolds, particularly for multivariate normal distributions. They focused on defining the mathematical properties and creating methods for the manifold, including sampling, projections, and probability density function calculations. Their work involved integrating existing statistical libraries like `scipy.stats` and implementing custom functions within the `geomstats` framework. The user also added unit tests to validate the behavior of the new statistical manifold.
gpu-programmingstatisticslie-groupsdeep-learningneural-networks
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.