A company originating from public research.

CartanLabs grew out of an observation first made in the laboratory and later encountered in several industrial settings: simulation campaigns often involve many closely related computations without fully exploiting the results already obtained.

The project began at Inria Saclay.

The research behind CartanLabs was conducted in a public research environment, focusing on numerical solvers and physics-constrained learning. The project is now in a maturation phase with support from Inria.

This public-research background is reflected in the way we develop and evaluate our methods, with particular attention to validation, reproducibility, and analysis of results.

Inria

French national research institute for digital science and technology.

Inria Saclay
Campus de l’École polytechnique
91120 Palaiseau

Our operating principles

01

Measure gains

Gains are evaluated on comparable cases, under the same computational conditions and against a reference method.

02

Document limitations

We also retain results that do not meet the expected criteria. They are part of the method's evaluation.

03

Protect your data

The infrastructure can be deployed within your computational environment so that geometries, parameters, and results remain under your control. Data processing is designed to comply with applicable data-protection requirements, including the GDPR.

Interested in this problem? Come work on it with us.

Numerical analysis, high-performance computing, machine learning, scientific software engineering.

Stay informed

Our technical updates and publications. No marketing emails.

We do not share your email address.

Explore
Home Technology About Careers Press & news Contact
Follow us
LinkedIn X
Legal
Legal notice Privacy policy Cookie policy Credits
Find us

Inria Saclay
École polytechnique campus
1 rue Honoré d’Estienne d’Orves
91120 Palaiseau, France

contact@cartanlabs.fr
© 2026 CartanLabs. All rights reserved.
Project supported in its maturation phase by Inria