For a better browsing experience and to benefit from all the features of credit-agricole.com, we advise you to use the Edge browser.
  • Text Size
  • Contrast

More than 500 participants and, in the end, three winning “models”... The first hackathon launched by Pacifica and CAA’s Data Science Academy, in partnership with the Ecole Normale Supérieure (ENS), proved a success! The presentation of the prizes of this data science competition, which consisted in challenging the modelling of the pure Fire premium of the agricultural multi-risk product, took place on 24 June at the Pacifica auditorium. Nearly a hundred people attended the event, in the presence of the three sponsors of the competition: Yann Renaut (DGA of Pacifica), Myriam Couillaud (DRH of CAA) and Marc-Antoine Hennel, Director of Digital Transformation at CAA.

Increase CAA’s visibility in the actuarial and data market

The winners of this first hackathon, which was open both externally (notably students in actuarial science and data science) and internally (individually or by team), are:

  • 1st prize: Mario Defrance (CAA) and Zacharie Buisson (Axa)
  • 2nd prize: Zakaria Bouchegoura (Axa)
  • 3rd prize: Elie Attali, Arthur de Rouck and Rémy Siahaan-Gensollen (ENSAE)

The target data to be modelled was based on the frequency of claims and the average cost of a claim, which made it possible to obtain the annual charge for a contract. The product pricing and reinsurance teams of Pacifica and the DataLab at BU International had previously designed two calculation models. The aim of this hackathon was therefore to challenge the results of these two models via machine learning solutions (artificial intelligence algorithms).

“We were impressed both by the number of participants and by the quality of the models presented to challenge our own models, which were based on varied approaches, from the simplest to the most complex (machine learning, etc.),” said Yann Renaut and Marc-Antoine Hennel during the award ceremony. “This hackathon helped to raise CAA’s profile in the actuarial and data science market,” said Myriam Couillaud. He has thus contributed to CAA’s good image and influence among schools of actuarial science, agronomy and engineers, with the aim of facilitating and recruiting future talents.”

 

What is Agricultural Multi-Risk?

As a reminder, Multirisk Agricole is an insurance product taken out by a farmer in order to secure his holding. It covers the exercise of his professional activity, the damage suffered by his operating buildings, the material stored and his financial and legal protection. The policyholder is thus covered effectively and over the long term so that, in the event of a setback, the continuity of his activity is ensured from both a material and a financial point of view. Multirisk Agricole generated nearly €1 billion in revenue in 2023 (source FFA), representing 36% of the entire French Agricole market. Fire risk modelling is a major challenge for this product since it represents a very large part of the overall load (35% in 2023, source FFA).

Associated Articles

AI
Devoxx: With AI, software engineering enters the framing era
29 Apr 2026
AI
Devoxx: Crédit Agricole combines AI, Clean Code and software quality
29 Apr 2026

If you wish to exercise your right to object to the processing of personal data for audience measurement purposes on our site via our service provider AT internet, click on refuse