Research Engineer
Scientific Computing · Computational Geometry · Deep Learning · Image-based Modelling · Computational Mechanics
I am a research engineer working at the intersection of numerical simulation, computational geometry, and deep learning for scientific imaging. I currently develop C++ and Python simulation software at CEA‑LIST, contributing smooth‑surface ray tracing acoustic propagation modules to the CIVA simulation platform, with applications to transcranial focused‑ultrasound therapy and industrial non‑destructive testing.
Over the past fifteen years I have developed scientific software across four countries, including from‑scratch finite‑element solvers, image‑based mesh generators, deep learning for X‑ray micro‑computed tomography, constitutive material laws, and, most recently, smooth B‑spline surface fitting combined with ray tracing. The underlying engineering profile has remained consistent: producing accurate model on which physics simulation can deliver the correct result.
The sections below summarise current and previous work.
Developer of the smooth‑surface fitting + ray tracing acoustic‑propagation module being integrated into the CIVA platform. The pipeline fits a Multilevel B‑spline (MBA) surface to processed CT‑segmented meshes, ray‑traces against it with a custom backend, and solves the multi‑surface Fermat path for ultrasound propagation.
A chronological tour of the positions and projects I have built and shipped.
LMPS / ENS Paris‑Saclay · Safran Aircraft Engines
Hyper‑elastic and elastoplastic material laws for explicit‑dynamics forming simulation of Safran fan‑blade‑class textile parts. Abaqus + LS‑DYNA HPC simulation, macro‑scale parameter identification, CT‑image deformation analysis.
HPCAbaqusLS‑DYNAVUMATPythonFortran
Mines ParisTech CMM (PSL) · Safran Aircraft Engines
Sole author of YarnPath, a regression‑CNN pipeline (TensorFlow / Python) that reconstructs yarn centrelines from µCT scans of 3D woven textile composites used in LEAP‑1A / LEAP‑1B aircraft engine root blades (A320neo, 737 MAX).
PythonTensorFlowAI / Deep learningµCT
Published in Composites Part A 186 (2024) 108396.
Ghent University · UGCT · Siemens PLM
TensorFlow‑based U‑Net µCT segmentation, image‑based meshing and FE modelling of carbon‑fibre 3D textile composites and short‑fibre‑reinforced polymers in collaboration with Siemens Industry Software. Synthetic image generation for training‑data augmentation, eliminating the manual‑annotation bottleneck.
PythonTensorFlowAI / Deep learningµCTHPCvox2tet
Institut Pprime
Custom image‑to‑mesh pipeline — Python version of vox2tet. µCT‑image‑based multi‑physics FE modelling of textile composites.
PythonMatlabFortranAbaqus
Karlsruhe Institute of Technology · Institute of Engineering Mechanics
From‑scratch FEM codebase for composite‑ microstructure optimisation, multi‑scale homogenisation, foam modelling, and CT‑image‑based FEM.
MatlabAbaqusHPC
Lviv National University · IAPMM NASU
From‑scratch FEM solver with 2D meshing, a posteriori adaptive refinement and exponentially-fitted Petrov-Galerkin for the convection‑diffusion problem.
C++FEMh-Adaptivityexponential splines
Publicly released code.
2018 — Present
C++ pipeline that converts labelled 3D voxel images into high‑quality conforming tetrahedral meshes for finite element simulation. Multi‑material marching cubes, smoothing, iterative remeshing, CDT/MMG/TetGen tetrahedralisation.
C++PythonCMakeEigenOpenMPlibtiffCDTMMGTetGen
github.com/skenop/vox2tet · Composite Structures 298 (2022) 116003
Full publication list: scholar.google.com/citations?user=HTOhzwsAAAAJ