Experience

Crownpeak Product Discovery

• United Kingdom
• IT Services and IT Consulting
• 1 - 100 Employee

• Senior Data Scientist

  • Sep 2019 - Present

  Making research and developing deep learning models for recommendation and search engines. Making research and developing deep learning models for recommendation and search engines.



Accor

• France
• Hospitality
• 700 & Above Employee

• Data Scientist

  • Apr 2016 - Sep 2019

  Project: Inside a team of Data Scientists and Data Engineers, development of an autonomous machine learning solution able to find optimal prices for all hotels owned by the company. I was hired as an intern by Accor during the 6 first months and hired by LumenAI as a consultant for Accor during the 15 next months. I'm currently hired by Accor as a Data Scientist. Realization: * Developing machine learning solutions to solve the endogeneity issue for markov decision process in continuous action space. * Developing fast optimization algorithm for non differentiable constraints problems. (Genetic Algorithms, Multi Agents Systems) * Developing deep learning architectures for multivariate times series problems. (MLP, CNN, RNN, MDN) * Developing distributed learning architecture on Amazon Web Services. (Docker, AWS Batch/ECS/ECR) * Developing anomaly detection and measurement algorithm for sequential data. (Machine Learning, Econometric theory) * Developing a simulator of the hotel market place. (Simulation Methods, Game Theory, Economics) * Developing internal Python libraries. * Take the lead for the Data Science and the Software Development part of the project. Technology: * Python, Spark, Tensorflow, Scikit-Learn, XgBoost, Hive, Kafka, MySQL, Docker, Git/GitLab, PyCharm, Sphinx, Linux, Hadoop, Amazon Web Services Show less



LumenAI

• France
• Software Development
• 1 - 100 Employee

• Data Scientist

  • Nov 2016 - Dec 2017

  I was consultant for AccorHotels during 14 months (see description above) and I have developed a Factorization Machine algorithm and some deep learning architectures for lumenAI. I was consultant for AccorHotels during 14 months (see description above) and I have developed a Factorization Machine algorithm and some deep learning architectures for lumenAI.



Pierre Fabre Group

• France
• Pharmaceutical Manufacturing
• 700 & Above Employee

• Statistician Economist Intern

  • Apr 2015 - Sep 2015

  Project: I have presented to the company a research project on Randomized Control Experiment applied to time varying market data which they have accepted. The goal of this internship was to develop a new experimental strategy to measure the return on investment of any Merchandising or Marketing campaigns realized by the company in its tracked pharmacies. Realization: * I have empirically proved the possibility to realize a precise causal experiment using Statistical Learning models and Econometric tools. * I helped the Merchandising department to deliver more robust analyses. Technology: * R * Stata Show less



Toulouse School of Economics

• France
• Higher Education
• 100 - 200 Employee

• Teaching Assistant in Microeconomics

  • Sep 2014 - Dec 2014

  I gave tutorials in Microeconomics to a class of 48 freshmen in Economy during 20 hours. The goal was to help them to solve exercises and better understand the course taught by the professor Philippe Alby. I gave tutorials in Microeconomics to a class of 48 freshmen in Economy during 20 hours. The goal was to help them to solve exercises and better understand the course taught by the professor Philippe Alby.



Pierre Fabre Group

• France
• Pharmaceutical Manufacturing
• 700 & Above Employee

• Economist Intern - International Strategic Coordination Department

  • Jun 2014 - Aug 2014

  The International Coordination Strategic department aims to coordinate all strategies of the countries where the company have an office with the headquarter vision. Mission and Realization: * Creation of a statistical report on country performances enabling improvement of the company visibility on worldwide performances. * Using Time Series modeling to predict the short run profit for all brands owned by the company. * Showing the interest for the company to use mathematical tools to help business decision makers. Technology: * Eviews * Excel Show less