About Me
Hi, my name’s Fabien and I am a physicist specializing in complex systems, with a strong passion for interdisciplinary innovation. My expertise lies in artificial intelligence and machine learning, which I apply to drive innovation in engineering solutions. Currently, I utilize my skills at Capgemini Engineering to design, simulate, and prototype advanced AI-driven systems that enhance efficiency and performance across various engineering domains.
I am most skilled in: Python and Modelization.
Projects
This is some of the most interesting research I have been able to conduct. (with my thesis, of course)
I started this project as a way if learning PyTorch and applied mathematics in neural network. I have learned many skills through this and been I’m very proud of having this in my portfolio.
Finding a way to mix animation art and math and can be a good way to put yourself in a mathematician's shoes when looking at shapes.
This project was carried out during the covid period after my thesis. It was a good way to get back to an old passion, drawing. Each of the animations illustrates fundamental Physics problems of complex systems.
Other personal projects
- Genome analysis: Using data from an ancestry test kit (MyHeritage, 23andme, etc.), for each nucliotide polymorphism, I extracted information from a genome database (Web Scraping).
- How does an artificial neural network understand an image? Animate step of a learning process that automatically annotated images caption and where each step of the process is represented (encoding by resnet and decoding by attention-lstm).
Experience
Matière et Systèmes Complexes Lab
PhD in Biophysics
September 2015 - September 2019
www.msc.univ-paris-diderot.frCharacterization of the NODAL signaling pathway during embryonic stem cell differentiation. It's a way to control cell differentiation trajectory. Management : Benoit Sorre and Pascal Hersen.
The aim of this work was to study the influence of the temporal stimulation of different morphogens on the expression of the “Nodal” locus and on the differentiation of stem cells. To control the temporal profile of the morphogen concentration sent to cells, we use an automated microfluidic cell culture chip that allows us to observe individual cell responses in real time with the fluorescence microscope. For this work, you need to have great skills in engineering, adaptability and image analysis.
Education
Established in 1808, the agrégation is a competitive examination for civil service in the French public education system.
During my learning of neural networks and machine learning, I took advantage of my year for a complete refresher in mathematics. I didn’t validate the competition, but this learning allowed me to be completely up to date on the theory of neural networks and machine learning.
Publications
Buckling along boundaries of elastic contrast as a mechanism for early vertebrate morphogenesis.
We have investigated the mechanism of formation of the body of a typical vertebrate, the chicken. We find that the body forms initially by folding at boundaries of stiffness contrast. These boundaries are dynamic lines, separating domains of different cell sizes, that are advected in a deterministic thin-film visco-elastic flow.
A Little More About Me
Alongside my interests in networks, research and engineering, some of my other interests and hobbies are:
- Creative Coding
- Build Educational Game See Here
- Hiking
- Cycling
- CrossFit
- Write essays about criticism of merit and of a form of Basic Income See Here and Here
- News release in MFRB See Here and Here.
Also, an overview when we mix code, art and mathematics
