Research Engineer - Postdoctoral Researcher Grenoble Conditional Generative Pde Surrogates For Ocean Model H/F

INRIA

Research Engineer / Postdoctoral Researcher @Grenoble: Conditional Generative PDE Surrogates for Ocean Model
Le descriptif de l'offre ci-dessous est en Anglais
Type de contrat : CDD

Niveau de diplôme exigé : Bac +5 ou équivalent

Fonction : Ingénieur scientifique contractuel

Niveau d'expérience souhaité : Jeune diplômé

A propos du centre ou de la direction fonctionnelle

The Centre Inria de l'Université de Grenoble groups together almost 600 people in 26 research teams and 9 research support departments.

Staff is present on three campuses in Grenoble, in close collaboration with other research and higher education institutions (Université Grenoble Alpes, CNRS, CEA, INRAE, ...), but also with key economic players in the area.

The Centre Inria de l'Université Grenoble Alpes is active in the fields of high-performance computing, verification and embedded systems, modeling of the environment at multiple levels, and data science and artificial intelligence. The center is a top-level scientific institute with an extensive network of international collaborations in Europe and the rest of the world.

Contexte et atouts du poste

The selected candidate will join the INRIA DataMove team (https://team.inria.fr/datamove), located in the IMAG building on the Saint-Martin-d'Hères campus (Université Grenoble Alpes), near Grenoble. The position involves close collaboration with the Institut des Géosciences de l'Environnement (IGE) (https://www.ige-grenoble.fr), also located on the same campus.

The contract can start as soon as possible (please allow 2-3 months for administrative processing) and will run until June 30, 2028.

DataMove and IGE offer a friendly, dynamic, and highly stimulating research environment, bringing together professors, researchers, and PhD and Master's students. Grenoble is an exceptional city surrounded by the Alps, offering a high quality of life and easy access to a wide range of outdoor activities (skiing, hiking, climbing, cycling).

Depending on the candidate's profile and career goals, this position can be offered either as a postdoctoral position (with a strong focus on publications) or as a research engineer position.

Advisors: Bruno Raffin (****@****.**) and Julien Le Sommer (****@****.**)

Mission confiée

Context

PDE surrogates are neural networks trained on data generated by traditional numerical PDE solvers. Their goal is to approximate these solvers at a significantly lower computational and memory cost. These approaches have recently attracted strong interest within the emerging fields of Scientific Machine Learning (SciML) and AI for Science.

Model architectures have rapidly evolved, from CNN-based designs to advanced approaches combining attention mechanisms, neural operators, and generative models. Recent examples include PDE-Transformer, Poseidon, and Universal Physics Transformer. In weather forecasting, several teams have reported breakthrough results using PDE surrogates, achieving near state-of-the-art accuracy at a fraction of the computational cost.

Deterministic PDE surrogates are typically trained by minimizing a mean squared error (MSE) loss between predictions and ground truth. However, they often suffer from a regression to the mean effect, which limits their ability to capture complex or chaotic dynamics.

Stochastic PDE surrogates address this limitation by incorporating generative modeling techniques such as diffusion models (DDPM), score-based models, or flow matching. These methods learn to transform a simple known distribution (typically Gaussian) into a complex target distribution using an iterative denoising process. At inference time, the model generates realistic samples conditioned on input data.

Compared to deterministic approaches, generative PDE surrogates better capture fine-scale structures and uncertainty, especially for chaotic systems. They naturally enable uncertainty quantification, making them well-suited for sensitivity analysis and inverse problems (e.g., parameter estimation via Bayesian inference).

NEMO (https://www.nemo-ocean.eu/) is a widely used ocean circulation model for research and operational forecasting in oceanography and climate science. It is based on the Navier-Stokes equations, coupled with a nonlinear equation of state linking temperature and salinity to fluid motion. Due to its turbulent and chaotic nature, uncertainty quantification is essential, motivating the use of stochastic PDE surrogates. Similarly, Croco () is an other ocean model specialized for costal and regional simulations.

The objective of this position is to design, train, and validate a conditional generative PDE surrogate for the NEMO and Croco models.

Our research

This project is a collaboration between IGE and the DataMove team, combining complementary expertise in ocean modeling and large-scale machine learning.

IGE is one of the leading contributors to the NEMO model and has deep expertise in ocean model numerical implementations, parameterizations, and applications. This knowledge is essential for data generation, validation, and physical interpretation.

DataMove has extensive experience in training PDE surrogates on large-scale supercomputing infrastructures. The team develops and maintains Melissa, an in-house platform (https://hal.science/hal-04102400v1 - ICML 2023), which enables efficient online training by streaming data directly from simulation runs to distributed multi-GPU training pipelines.

Melissa also supports active learning strategies, allowing simulations to focus on challenging regimes and thereby improve both model quality and training efficiency (https://hal.science/hal-04712480v1).

Principales activités

The first objective is to become familiar with the scientific context, including PDE surrogates and ocean modeling.

The second phase will focus on developing hands-on expertise in training PDE surrogates using benchmark PDE systems and standard architectures such as U-Net, FNO, and related models. Existing workflows are already available within the team to support this phase.

The core of the project will then consist of designing and training a stochastic surrogate model for ocean simulations. The target approach involves a generative architecture operating in latent space. Both training from scratch and fine-tuning of existing foundation models for PDEs will be considered.

You will lead this work in close collaboration with experts from both DataMove and IGE. The project benefits from regular meetings, continuous interaction, and strong team support-you will not be working in isolation. You will also have access to state-of-the-art supercomputing resources with high-end GPUs.

Compétences

We are looking for a candidate with strong skills in deep learning (e.g., transformers, generative models), solid background in PDEs (CFD is a plus), and good programming abilities in Python for ML/DL development.

The ideal candidate is curious, proactive, and enjoys numerical experimentation, with a strong motivation to apply cutting-edge AI techniques to geoscience problems.

Candidates should hold:
- A Master's degree (or equivalent) in computer science or a related field for the research engineer position.
- A PhD for the postdoctoral position.

Technical skills in Linux environments, strong Python development practices, and familiarity with C/C++ are highly appreciated. Experience with modern development tools and workflows (git, CI/CD, package managers such as conda/nix/guix/uv) is a plus.

A good level of written and spoken English is required, as we are an international research team and English is our working language.

To apply, please submit your CV, references, academic transcripts, and (if available) your Master's or PhD thesis manuscript. You are also encouraged to include any additional material that demonstrates your skills (e.g., GitHub projects, code samples). Please provide contact details for referees who can comment on your work and qualifications.

Avantages

- Subsidizedmeals
- Partial reimbursement of public transport costs
- Leave: 7 weeks of annual leave + 10 extra days off due to RTT (statutory reduction in working hours) + possibility of exceptional leave (sick children, moving home, etc.)
- Possibility of teleworking (90 days / year) and flexible organization of working hours
- Social, cultural and sports events and activities
- Access to vocational training
- Social security coverage under conditions

Rémunération

From 2,692 € (depending on experience and qualifications).