We support successful tech founders on their journey to innovate and scale.

- HR: automated profile discovery within the organisation to improve internal mobility- Electric Vehicule: identified thermic vehicule drivers with driving habits compatible with electric cars by leveraging "connected vehicule" data- Supply Chain: monitored production margins, in a context of semi-conductors' shortage, to help prioritize the production of cars with the highest margins- Manufacturing: automated manufacturing analysis to help factories to optimise their processes and improve the quality-

Help fight Covid by bringing contact tracing within all organisations while preserving employees data privacy.

- AML/CFT: fraude detection app with a semi-supervised in-house solution based on Kohonen maps and graph metrics - AML/CFT: high performance custom screening solution for watchlists - Credit scoring: complete redesign of the FIBEN credit score based on neural network - Search engine for an heterogenous corpus of legal documents - Banking supervision: AI augmented dashboard to help auditors analyze the data received from banks

Worked for Banque de France on : - dashboarding for suspicious transactions with Tableau - web scraping of booking.com, hotels.com and tripadvisor.com using Scrapy - Credit scoring

Detecting adwares and malwares using NLP methods:- Identify malware families based on VirusTotal API's output on files SHA256- Plateform for cyber analysts to find the most likely name of each malware as specified by different vendors on VirusTotal

Developed a computer vision algorithm to detect cancer cells in infrared images using OpenCV and ML algorithms which helped trigger a seed-round fundraising.

Main developer under Arnaud de Bruyn, the project manager, I was in charge of automating the marketing analytics analysis on the platform using R: - segmentation - conjoint - positioning - predictive - sentiment analysisThe main skills were: - Data visualisation - Statistics

I worked closely with Prof. Arnaud De Bruyn. The research purpose was to augment data in programmatic campaigns through the use of weather data. In order to improve the predictive performance of the models, I learned Machine Learning algorithms like Random Forest, SVM and KNN.

2 years apprenticeship at 1000MercisFirst half : eCRM campaigns for P&G and DanoneSecond half : Real Time Bidding campaigns (RTB) for multiple clients from the insurance to the retailing sectors in an innovative and ever changing environment.