Decisions must invariably be made when one or more key quantities are uncertain.

We offer expertise in the analysis and modeling of ecosystem data.

We are developing a unique collaborative framework to review and share the projects and streamline knowledge between stakeholders.

Our objective is to facilitate the modeling of conflicting information, to quantify the uncertainties and to elicitate expert knowledge to make a decision.

A set of Recipes in Data-Science


We are developing and using Agent-Based Models for exploring population movement, resistance outbreaks and invasive species potential.

Besides standard (eco)toxicity statistics, we have a strong expertise for higher tier analysis, including several landscape-level models.

We have develop a landscape software (briskaR) used by the EFSA to support the assessment of GM impact on Lepidoptera (briskaR-NTL).

We are user of spatial mechanistic modelling in julia (DynamicGrid.jl & Circuitscape.jl) and contribute to their development (see this paper).


We develop tools to facilitate the collection and formating of spatial data (see picture on the right).

We have a strong expertise in GIS data process (QGIS, GDAL, …).

We can provide support to develop a Bayesian model for spatialized data though GLM, GLMM).

We can export our works in GUI interface using Leaflet or QGIS plugin.


We have a strong experience in Bayesian inference in food web ecology and in ecotoxicology where we developped and maintained several softwares (morse, rbioacc, rPBK).

We provide expertise in image & video treatment and analysis (openCV, Machine Learning).

We adapted the very new machine learning Neural ODEs coupling Deep Neural Network with Ordinary Differential Equations to the field of ecotoxicology.

We can also support you on more traditional statistics like multi-dimensional data analysis (PCA, LDA, CCA, FA).


We have a strong expertise in the environmental risk assessment of chemical products (scientific reportstipping points).

We can help you to optimise your data analyses workflow with standard statistics (NOEC, LOEC, LCx, ECx, SSD, MFx),

We have used and developped more advanced modeling techniques (DEB models, BioAccumulation, GUTS, TK, TK-TD, PB-TK, Reproduction, Growth) (see our packages: morserbioaccrPBK).

Moreover, we have specific expertise in probabilistic programming (i.e. Bayesian inference) of (eco)toxicological models, enabling the fine-tuning of mechanistic models with lab or ground data.


We develop models in R, Python, Julia or C++.

We can work with various GUI technologies (html / js, Shiny, Jupyter, …) .

We are experienced in standard software environments (docker, kubernetes, git, gitlab/github, …).

We can work on embedded systems such as custom acquisition setups (C, C++, standalone / linux).

Delivering within a collaborative plateform

we deliver our work within a home-made
user-friendly collaborative platform

Ecological studies often involve many different scientific topics, different jobs, different organization.
We believe communication is key in such situations, and that collaborative tools can make it easier.
To do so, the platform is built around huge knowledge base including relevant regulatory texts, model summaries and scientific references. We want each result report to be crystal clear, with a complete scientific legitimacy.

Collaborative Workflow

With its comment and discussion features, we want our platform to ease the work between different teams and specialities. Of course, your work privacy is a primary concern to us, and you can choose what to share with whom.

Exploring Scenarios

The platform allows users to navigate through different scenarios or to run new simulations by easily changing model parameters. It makes it easy to add new data to a current model or simply test new datasets.

Project Management

Every change on your project (data, models, comments) is stored and can easily be viewed. Moreover, we use software “containerization” techniques, to make sure you will be able to re-run the project code in the future.

Cloud Computing

Cloud computing is great for IT resources management. Whether your model needs a minimal computer or a high amount of RAM and CPUs/GPUs, we can select the appropriate hardware only for the time you will use it.

Decision Support

In order to optimize the communication of your studies, we provide a set of tools for viewing (graphics and maps) and formatting information (organization of text and graphics). By providing hierarchical and contextualized information, we help to be didactic.

Want to Partner with an Ecological Modeler and a Computer Scientist?

Virgile Baudrot

Co-Founder, CEO

PhD in Ecological Modeling

Thomas Kleiber

Co-Founder, CTO

Computer System Engineer

Contact Us

    Our Partners

    Entreprise soutenue par IMT Mines Alès dans le cadre de son incubateur, par l’Europe, la Région Occitanie et Alès Agglomération.

    Design 2022 by Qonfluens

    34190 Saint Bauzille de Putois