Merci beaucoup d'avoir accepté l'interview.
Can you please describe your journey towards a PhD?
My love for science, particularly maths and physics, has shaped my academic journey from the beginning. While I initially considered becoming an engineer to apply science in a practical context, I soon realised that an academic career would allow me to stay even closer to daily scientific aspects. This realisation pushed me to pursue a career in research.
I joined the École Normale Supérieure Paris-Saclay (ENS Cachan at the time) to get training in research, majoring in mechanics and later specialising in computational mechanics. My research experiences in Dublin and Paris confirmed my taste for both research and computational mechanics.
In my final year at ENS, I reached out to a professor working in computational mechanics to discuss PhD opportunities. He was just finishing developing a project in partnership with the CEA. After he introduced me to the subject, we quickly decided to work together, and that's how I started my PhD.
Can you please briefly describe your PhD thesis work?
This PhD was conducted in collaboration between the CEA and the LMPS, formerly known as LMT, which is the mechanics laboratory of ENS Paris-Saclay and CentraleSupélec. The objective of this research was to develop reduced-order modelling techniques to predict the probability of failure of power plants subjected to earthquakes. Predicting such probabilities requires performing numerous non-linear computations for various earthquake magnitudes at various stages of the structure's life. Each computation is numerically intensive, and calculating all probabilities is therefore computationally expensive.
To address this challenge, the goal was, first, to develop methods that reduce the computational burden of individual non-linear simulations, projecting the partial derivative equations describing the mechanical behaviour of the structure onto a reduced-order basis learnt on the fly.
Then, the goal was to develop a global methodology that benefits from the redundancy between computations to further decrease the overall cost of these stochastic studies.
What did the thesis bring you?
Undertaking a PhD is a great opportunity to deeply explore a subject you are passionate about. Over this three-year journey, I learnt what it means to conduct scientific research, developing a sense of critical thinking and a rigorous scientific approach.
One of the most significant aspects of the PhD experience is the development of independence. Unlike in Master's programs, where problems often have known solutions, a PhD involves tackling questions for which no prior answers exist. This requires taking initiative, proposing novel approaches, and positioning your work within the context of a vast and evolving body of literature. These challenges enable intellectual growth and refine your ability to contribute meaningfully to your field. The process is deeply rewarding. Discovering new connections between existing knowledge or identifying innovative solutions brings immense satisfaction.
Of course, there are challenging moments when research inevitably stagnates, and you don't come up with any new ideas for several days or even weeks. For me, however, doing a PhD remained a very good experience that only confirmed my enthusiasm for research.
What are you currently working on?
I find that doing research is a very fulfilling way to satisfy my thirst for knowledge. The opportunity to learn something new every day is a wonderful feeling, and it has reinforced my desire to stay in academia. To pursue this goal, I joined École Polytechnique for a postdoctoral position.
The ultimate aim of my postdoctoral research is to achieve individualisation and real-time simulation of a lung digital twin using model-order reduction techniques. This work focuses on translating theoretical tools into clinical applications for patients with idiopathic pulmonary fibrosis (IPF), which I find fascinating for several reasons. First, the societal motivation behind the research is clear: the problems we are addressing come directly from doctors and the clinical environment, making it highly motivating to find solutions. Second, from a scientific perspective, this subject is very interesting. Modelling soft tissues involves working within a non-linear finite strain framework, which is a challenging area of mechanics. Finally, during my PhD, I fell in love with reduced-order modelling, which uses tensor decomposition to represent solutions to physics problems in a way that significantly reduces computational cost. The medical field’s growing need for accurate and efficient digital twins provides the perfect application for these techniques. This area presents numerous scientific challenges yet to be addressed, making it an exciting opportunity to push the boundaries of reduced-order modelling for non-linear computational mechanics.
How do you benefit from your PhD today?
This postdoc position illustrates how the knowledge and skills developed during a PhD can be applied to fields beyond the original focus. I rely on the theoretical tools I discovered and worked with during my PhD to tackle challenges in a new area. With a solid foundation in reduced-order modelling, I have had the opportunity to explore the fascinating world of biomechanics and finite strain. This transition has allowed me to apply the scientific method I learned during my PhD to new applications, broadening my expertise.
