Who am I?

Passionate about mathematics and computer science, I work on making both worlds meet and get the best out of each other. I have had and still am having a dual academic and engineering career in which I let industrial questions drive my research interest and let my engineering problems benefit from current research. In parallel to research and industrial problem solving, I mentor students in their learning journeys.
I am currently interested in Algorithmic Differentiation and Signal processing (especially interactions between engineering and compressed sensing), having defended my habilitation to supervise research at the Université Nice Côte d'Azur in December 2025 on these subjects.

A Brief History of me

Here are the key highlights from my career.

• Currently Advanced Researcher
  As of January 2026, I am currently part of the GammaO team within the INRIA Centre of Saclay. My research evolves around Algorithmic differentiation and source transformation; a paradigm in which a piece of software code is analysed and another piece of code is generated in the same language, implementing its derivative. In particular, my developments are made available through the open source software Tapenade. Prior to joining Saclay and GammaO, I was part of the ECUADOR team at INRIA Sophia Antipolis. (Jan. 2024-Dec. 2025)
• Currently freelance Mentor/Teacher
  Contributed to London School of Economics online programs. Teaching mathematics and calculus.
  Mentoring students via Open classrooms in their data/AI projects.
• Expert engineer
  Artelys France, Paris.
  Industrial software development for power systems and optimization.
  Working with RTE on the open source projecs dynawo and powsybl.
  Development of a robust and combinatorial optimization using HPC and knitro.
  Contributions to the optimization engine of the suite crystal of software.
  (Some) Technologies used: C++, Python, Pandas, Docker, Airflow, Sun grid engine, AWS.
• Algorithm engineer
  Continental UK, Burgess Hill.
  Development of camera/radar sensor fusion algorithms.
  Development of calibration algorithms for multicamera systems in autonomous vehicles.
  (Some) Technologies used: C++, Matlab, Numerical optimisation.
• Assistant professor of mathematics
  School of mathematics and statistics.
  Beijinjg Institute of Technology.
  Teaching applied mathematics (optimization, regression, matrix analysis, computational science engineering).
  Researching signal processing (compressed sensing, stochastic PDEs).
• Postdoc
  Within the Chair for mathematics C (Analysis) at RWTH Aachen.
  Within the Department of Mathematics of Drexel University.
  Doing undergraduate teaching (calculus/algebra/numerical analysis) and graduate teaching (Mathematical and computational genomics / Mathematical foundations of machine learning)
  Researching compressed sensing and its applications. (to PDEs, MRImages, Metagenomic data survey)

On my free time, I tend to fiddle around with computer vision an Raspberry Pi's and similar nerdy things. Or I am out playing music or running marathons or hiking or... Should you be so curious, you may have a look at a more extended academic resume, or check my Linkedin profile. Why would one want to inflict this upon themselves beats me.

Some key points

I am a fervent believer in Open science and reproducible research. As such most of my work and all of my recent works can be found online on various platforms: my research page; my list of publications; my software on GitHub; some recent papers on arXiv. You may also drop me an email, should you need more references. I have also contributed to some open source projects, such as

• Tapenade: a tool for algorithmic differentiation of C and ForTran codes. In particular I have already improved some aspects of tapenade in its handling of function pointers, pointer analysis, profiling of checkpointing...
• dynawo: a suite of simulation tools for power systems. In particular I work on automatically controling voltage on a (very) large high voltage network.
• powsybl: PoWerSystemBlocks. In particular I have implemented a few functions regarding remedial actions and passed on some features to the python API: PyPowSyBl.
• CSPDE: Compressed Sensing for PDEs. This is pretty much all in-house and is a tool to compute numerical solutions to parametric PDEs through generalized polynomial chaos.

I am very grateful to have had the opportunity to teach and talk in various places around the world, including:

• Southeast Asian Mathematical Society School on Modern Trends in Signal and Data Processing which took place in December 2022 in Manila.
• High performance AI in the industry within the Teratec forum in June 2022. I talked about robust and stochastic combinatorial optimization on an HPC.
• Gene Golub SIAM Summer School on the Theory and Practice of Deep Learning which took place (virtually online) in South Africa in July 2021.
• Summer program on Mathematical Signal Processing and Data Analysis within the Academia Sinica in Taipei in Summer 2019.
• Polytech'Sophia (High dimensional linear systems; Introduction to machine learning in python)
• London School of Economics (Calculus)
• Beijing Institute of Technology (Matrix analysis, Computational science engineering, Modern optimization, Modern regression)
• RWTH Aachen University (Maths 1/2/3, Mathematics of machine learning, Mathematics of computational genomics)
• Drexel University (Introductory analysis, Differential equations for engineers, Numerical analysis 1 & 2)

As a researcher, I am generally interested in applications of sparse approximation and signal processing to general mathematical and engineering problems. I am now tackling new challenges in algorithmic differentiation and contribute on theoretical aspects as well as implementations within the source to source differentiation software TAPENADE. I have contributed to many very diverse areas including (in non logical order) Optimal Coherence Tomography; Numerical solutions to parametric PDEs; Numerical recovery of sparse signals; Metagenomics; ...

This page is an ongoing work and contains only a subset of what it is hoped it will contain soon