I am a PhD-level consultant designing, implementing and optimizing bespoke mathematical models, algorithms and numerical simulations. My ten years of international experience at the crossroads of applied mathematics and programming allow me to find the algorithm best suited to your situation, or to optimize codes that crunch a lot of numbers to improve speed, reliability and accuracy (cf. http://mathieu.bouville.name/en/science/numerical-programming.html).
I have worked in fields as varied as materials science and applied physics; investing, personal finance and financial planning; renewable energy and energy savings; sport statistics (and I am always interested in working in new domains).
I have a doctorate degree from the University of Michigan. I co-authored 15 scientific articles on mathematical modeling and numerical simulations, including two as first author in Physical Review Letters.
Mathematical modeling: An introduction — Mathematical modeling means taking a concrete problem, and solving it through math, be it by solving equations or developing custom software doing calculations. Modeling may replace experiments or complement them.
A player, a computer and a mathematician try to solve a sudoku puzzle — A sudoku puzzle can be solved by deduction, by brute force or by turning it into a different problem, depending on whether you are a player, a computer or a mathematician.
Monte Carlo simulations: from physics to financial planning — Monte Carlo simulations have diverse applications in physics, materials science, investing, etc. They are used in situations where randomness plays a part (or which can be treated as such).
How to implement calculations efficiently and accurately — Programmers know how to improve the speed and robustness of their codes. However, when a program does a lot of calculations, new problems arise and new solutions are needed.
At the University of Cambridge, I developed a new, faster (factor 100) computer simulation method within the framework of Monte Carlo lattice model to predict polymer glass-transition temperatures.
In order to better control dopant diffusion during device processing, I studied the effect of stress on point defects in semiconductors in collaboration with mechanical engineers.
Working closely with experimentalists, I designed a mathematical model to determine under what conditions pits would form in heteroepitaxial semiconductor thin films. [Phys. Rev. B 70, 235312 (2004)]
On a project I initiated, I showed how various parameters affect the interplay between martensite and pearlite formation in steel (which cannot be done experimentally). [Phys. Rev. Lett. 97, 055701 (2006)]
Collaborating with experimentalists, I used phase-field simulations to explain how alloying may stabilize polycrystalline thin films and dramatically enhance their performance. [Phys. Rev. Lett. 98, 085503 (2007)]