Research topicsI was PhD candidate at the Jean Kuntzmann laboratory under the supervision of Georges-Henri Cottet, Christophe Picard and Iraj Mortazavi on the subject of numerical modelling and high performance computing for sediment flows. After having obtained my PhD in numerical simulation I moved to the field of computational geometry. I am currently completing my postdoctoral research under the supervision of Boris Thibert and Quentin Mérigot in the field of anidolic optics. The common denominator between all my research is high performance computing and its application to solve compute-intensive problems.
I am still occasionally contributing to HySoP (Hybrid Simulation with Particles), a Python library dedicated to high performance distributed flow simulation based on semi-Lagrangian particle methods and targeting heterogeneous architectures. This library is built on top of various computing backends including C++, Fortran and OpenCL and will soon be available under an open source Apache License Version 2.0 (APLv2).
- Design of optical components by solving inverse problems.
- Optimization methods coupled with global cost functions.
- Implementation of efficient distributed multiscale approaches.
- Hardware accelerated differentiable raytracing.
Computational Fluid Dynamics (CFD)
- Direct Numerical Simulation (DNS) of turbulent incompressible fluid dynamics.
- Fluid coupling and interaction with sediment-like species.
- High order semi-Lagrangian methods (remeshed particle methods).
- Finite differences, spectral and pseudo-spectral methods on Cartesian grids.
- Vortex penalization method.
High Performance Computing (HPC)
- Multi and manycores architectures, coprocessors (CPU, GPU, MIC).
- Distributed hybrid CPU/GPU computing with MPI.
- JIT Cuda/OpenCL code generation from symbolic expressions.
- Runtime code and run parameters autotuning depending on the target architecture.
- Design and optimization of FPGA-based computing systems with OpenCL.
Other topics of interestApart from my expertise in GPU computing and my interests in computational fluid dynamics and high performance computing, I have a taste for:
- Numerical simulation and optimization
- Image reconstruction, medical imaging
- Scientific visualization, 3D graphics rendering
- Optimal transport and its applications
- Signal processing and wavelets
- Proof of work / hashing algorithms