Research Engineer Position at CNRM in artificial intelligence for weather forecasting F/M

As a public weather and climate expert, Météo-France is at your side to contribute to your day-to-day safety and help you make the best decisions in a changing climate. Faced with dangerous weather episodes that are even more intense and more frequent as a result of climate change, our missions in the service of your safety are crucial. We mobilize our expertise and our scientific and technological excellence to help you anticipate and adapt to challenging weather and climate phenomena.

Find us online: https://meteofrance.com/carte-didentite-de-meteo-france

The DESR brings together the research entities of Météo-France (mainly CNRM, SAFIRE and LACy), the Ecole Nationale de la Météorologie and the shared administrative and IT support services (PGA).

The CNRM is a Joint Research Unit (UMR 3589) under the joint supervision of Météo-France and CNRS. The CNRM conducts research in the field of meteorology and climate, from the observation, understanding and modelling of processes to the development of weather forecasting and climate projection systems that can be transferred to Météo- France's operational services.

The GMAP (Groupe de Modélisation et d'Assimilation pour la Prévision - Modelling and Assimilation Group for Forecasting), based in Toulouse, is one of the CNRM's 6 research groups or centres. Its mission is to maintain and develop Météo-France's operational numerical weather prediction (NWP) systems and to conduct research in preparation for future versions.

Why join us?

Embark on a stimulating adventure in the service of all, alongside men and women committed to the daily challenges posed to our society by the weather and climate. And enjoy the following benefits: flexible working hours, RTT, telecommuting, administrative restaurant or luncheon voucher, 75% contribution to public transport, contribution to health insurance, sports and cultural associations depending on the site (climbing, gym, pottery, theater, etc.).

Other benefits await you, come and discover them!

This position is part of the WeatherGenerator european project (https://weathergenerator.eu/) led by ECMWF. The objective of the WeatherGenerator is to improve weather and climate predictions, enhance renewable energy planning, and provide better tools for mitigating the effect of extreme events like floods and heatwaves. It combines the latest machine learning with high performance computing, creating a robust tool for scientific discovery and real-world applications. Building on the key concepts of large-scale representation learning, WeatherGenerator is meant to be a foundation model, trained on multiple datasets, from reanalyses to satellite observations. The resulting model will be accessible through finetuning for a range of applications in the area of Earth system science. Meteo-France is taking part in this international endeavour and aims at applying the WeatherGenerator framework to regional weather forecasting, with a focus on the accurate modelling of under-meso-scale extreme events over metropolitan France.

Building on the WeatherGenerator model, the aim of the present position is to evaluate its capabilities for high-resolution weather forecasting over western Europe. The main objectives are the following :

• Fine-tune the foundation model on Météo-France Arome data ;

• Evaluate its performances, and benchmark against other weather forecasting models developed by

Meteo-France AI-weather forecasting research team ;

• Implement and evaluate strategies to use the WeatherGenerator in a probabilistic forecasting setting ;

• Work collaboratively with the international project partners ;

• Contribute to workshops and hackathons organized by the projet, and more generally take part to dissemination activities (including the production of project deliverables and scientific publications).

Additional information

This position is open to people with disabilities.

Office hours.

Fixed-term contract of 18 months.

Remote work available: 1 day a week maximum.

The ideal candidate would have the following qualifications :

- A master degree in atmospheric sciences, statistics or artificial intelligence ; a PhD is an advantage

- A strong background in deep learning algorithms, in particular transformer networks and deep generative

models

- Experience in geophysical problems would be appreciated, at least a strong interest for applied research

in atmospheric physics is highly recommended

- Proficiency with Python programming and AI librairies (PyTorch)

- Experience with processing large volumes of data (distributed dataloading and computing frameworks)

- Experience of working in a Linux-based environment

-Level in french not required

- Aptitude for scientific work, written and oral communication in English, meetings abroad are envisioned

Application deadline: 01/11/2026

Email addresses for further information: laure.raynaud@meteo.fr , clement.brochet@meteo.fr, matteo.salis@meteo.fr, sara.akodad@meteo.fr

- A scientific curiosity, autonomy, rigor in the interpretation of the results