Postdoctoral position in Hybrid Modelling and Optimal Control of Microalgae
Inria - Centre d’Université Côte d’Azur (GREENOWL) / DareWin EvolutionPost-Doctoral Research Visit F/M – Postdoctoral position in Hybrid Modelling and Optimal Control of Microalgae
Nice – Alpes Maritimes.
- Type of contract: CDD – 18 months.
- Required qualification level: PhD.
- Function: Post-Doctorant.
Where
The Inria Centre at Université Côte d’Azur includes 42 project teams and 9 support services. The centre’s staff (about 500 people) is made up of scientists of different nationalities, engineers, technicians and administrative staff. The teams are mainly located on the campus in Sophia Antipolis and Nice, as well as in Montpellier, Bologna and Thessalonica. The Inria Centre at Université Côte d’Azur is a major player in terms of scientific excellence through its results and collaborations at both European and international levels.
Background and Benefits of the Role
GREENOWL team
The joint project-team GREENOWL (Inria, CNRS, Sorbonne Université) jointly with LOV (Villefranche-sur-Mer) is led by Olivier Bernard. Its main objective is to assess the resilience and adaptation capacity of microbial ecosystems facing climate change, and to master the directed evolution of microorganisms to develop new sources of energy, proteins, and high-value molecules. To achieve this, GREENOWL deploys theoretical tools in control theory, artificial intelligence, and metabolic simulation, validated experimentally through dedicated platforms and industrial partnerships.
The EPIX Project
Long-chain omega-3 fatty acids, particularly EPA (eicosapentaenoic acid), are essential for human health and currently rely heavily on overexploited marine fish stocks. Microalgae represent a sustainable alternative, and mixotrophic cultivation combining photosynthesis and heterotrophic growth on organic carbon (glucose) holds great promise for boosting productivity.
Supported by the IdEx Université Côte d’Azur (“Start-up DeepTech – Partenariat Innovation” program), the EPIX project is a close collaboration between GREENOWL and the DeepTech startup DareWin Evolution, which specializes in dynamic Darwinian selection of microalgae without genetic modification. The project aims to develop digital twin and optimal control strategies to drive the accelerated selection of a mixotrophic or autotrophic microalgal strain overproducing EPA.
Assigned Task
The postdoctoral researcher will be in charge of the mathematical modelling and optimal control work packages of the EPIX project, working in close interaction with Olivier Bernard, Francesca Casagli, Walid Djema (GREENOWL), and the R&D team of DareWin Evolution (Hubert Bonnefond, Charlotte Gaviard).
The main objectives of the mission are to develop and validate a digital twin of mixotrophic or autotrophic growth and lipids production:
- Combine mechanistic models (photosynthesis, mitochondrial respiration, intracellular lipid dynamics, and metabolic models) with Artificial Neural Networks (physics/metabolism-informed machine learning) to capture complex non-linearities arising from the simultaneous co-utilization of light and organic carbon, while guaranteeing stoichiometric constraints and state positivity.
- Design optimal control strategies for Darwinian selection: Analyze the mathematical model (or a simplified surrogate model) using Pontryagin’s Maximum Principle (PMP) and design Model Predictive Control (MPC) and online learning algorithms. These algorithms will compute optimal dynamic environmental stress trajectories (oxidative stress/ROS via light modulation, temperature shifts, organic carbon pulsing, light/dark cycles) to separate strains in minimal time and favor the emergence of EPA-rich phenotypes.
- Implement the designed strategies on an experimental platform: Using the ODIN+ framework developed at Inria, implement the control strategy and contribute to the experiments to get data for calibrating the models and validating the control strategies.
Main Activities
- Formulate and calibrate mechanistic and metabolic models of mixotrophic microalgal growth and lipid/EPA synthesis using existing datasets and new experimental data generated by DareWin Evolution.
- Develop hybrid modelling architectures integrating neural networks with metabolic stoichiometry constraints.
- Perform optimal control analysis and develop MPC algorithms for minimal-time strain separation in a Selectiostat.
- Implement the predictive control and online adaptation algorithms as operational plugins within the Inria proprietary supervision software ODIN+.
- Participate in experimental campaign (30% of the time) follow-ups with DareWin Evolution and publish methodological results in top-tier control, machine learning, and biotechnology journals.
Skills
- Education: PhD degree in Applied Mathematics, Control Theory (Automatic Control), Bioprocess Engineering, Computational Biology, or a related field.
- Scientific skills: Strong background in dynamical systems, mathematical modelling of biological/metabolic systems, and optimal control (PMP, MPC). Experience with hybrid modelling / Scientific Machine Learning (PINNs, neural ODEs, constrained ML) is a strong asset. Knowledge of biological systems.
- Programming: Proficiency in scientific computing using Python (JAX, PyTorch, SciPy) and/or MATLAB.
- Languages & Soft skills: Fluency in English (written and spoken); knowledge of French is a plus for daily interactions with the startup partner. Strong interest in experimental validation and multidisciplinary teamwork at the interface of mathematics and biotechnology.
General Information
- Remuneration: According to the candidate profile (Inria postdoctoral salary scale).
- City: Mainly Sophia Antipolis (Valbonne) – experimental part at Laboratory of Villefranche sur mer.
- Centre Inria: Centre Inria d’Université Côte d’Azur.
- Start: As soon as possible.
- Duration: 18 months.
- Deadline for application: 15/12/2026.
- Contact: job@darewinevolution.com (Subject of the email: POSTDOC 18M INRIA).
More information: https://www.linkedin.com/posts/hubert-bonnefond-67774aa5_postdoc-position-microalgae-inria-darewin-ugcPost-7528466018692882432-ROJK/