Serge-Henri CACHIA, PhD

Simulations du flux sanguin au niveau des valves aortiques avec interaction fluide/structure (CFD, FEA, IMGA). Numerical simulations of blood flow near aortic valves (CFM/FEA, IMGA, fluid structure interaction). Python, Rhino3D.

Aider les actuaires à mettre en place un outil de calcul de métriques (Python) à partir d’un outil existant sur Excel et de nouvelles règles métier répondant aux normes internationales.

Accompagner des équipes dans la mise en place de pipeline de déploiement jusqu’à la mise en production avec XL release.

Industrialiser des pocs de maintenance predictive. Développer des APIs Flask, ETL. Dashboarding KPIs calculés a partir de données d’un outil de ticketing (Service Now). Industrialise data science POCs. Develop Flask APIs. ETL. Dashboarding KPIs calculated thanks to data from a ticketing tool (Service Now).

Développer un poc sur une méthode innovante d'inventaire de jeu de cartes à collectionner basée sur la reconnaissance d'image. The aim of this mission was to research and develop an innovative inventory methodology for a trading card game based on image recognition. Web scrapping [Beautiful Soup], logistic regression, CNN (keras) at [py]tesseract (ocr).

The aim of this work was to coordinate a team of volunteer entrepreneurs in Chengdu, China, to organize a networking between Western and Chinese companies in the city. I also had the opportunity in this period to provide help to these entrepreneurs to optimize the processes in their business (especially high quality fast food business).

Thesis title: Molecular dynamics simulations of water transport properties and magnetic resonance relaxation in cement nanopores Water transport properties in cement are important for the cement industry. At the nanoscale, a nondestructive experimental method, 1H nuclear magnetic resonance [NMR] relaxometry, can be used to quantify these properties. However previous results have proven difficult to reconcile with the understanding of cement. The purpose of this work was to use molecular dynamics [MD] simulations to better understand water dynamic properties in cement and hence to better interpret some of the NMR data. This work was part of the TRANSCEND european project composed of 15 PhD students around Europe and financed by Marie Skłodowska-Curie Actions. Technologies used: FORTRAN, C++, DL_POLY (MD) software

This internship purpose was to understand in a better way how a carbochloration reactor worked and which input parameters could be changed in order to optimize the reaction. This was performed thanks to numerical simulations and the analysis of the output data.