Experience

Huawei Paris Research Center

• France
• Telecommunications
• 1 - 100 Employee

• PHD Student

  • Apr 2022 - Present

  Boulogne-Billancourt, Île-de-France, France As a PhD candidate, my research primarily focuses on time series analysis, specifically classification and forecasting. In the early stages of my thesis, I explored data augmentation for time series and some methods for synthetic time series generation, addressing challenges associated with high labeling and cleaning costs. I expanded my research to adversarial attacks, constructing innovative attack frameworks while formulating adaptive defense mechanisms. In addition, I am… Show more As a PhD candidate, my research primarily focuses on time series analysis, specifically classification and forecasting. In the early stages of my thesis, I explored data augmentation for time series and some methods for synthetic time series generation, addressing challenges associated with high labeling and cleaning costs. I expanded my research to adversarial attacks, constructing innovative attack frameworks while formulating adaptive defense mechanisms. In addition, I am working on time series forecasting, transport optimal theory, and influence functions for data selection. Although my work is centered around time series classification and forecasting, the methodologies developed can be readily adapted to other domains such as computer vision or NLP.

• AI Research Engineer

  • Dec 2021 - Apr 2022

  Boulogne-Billancourt, Île-de-France, France In this fixed-term contract, I devoted my research efforts to data augmentation and generation, specifically focusing on state-of-the-art methodologies. I applied these techniques to the field of failure prediction, a domain frequently confronted with imbalanced data.



SNCF

• Rail Transportation
• 1 - 100 Employee

• Data Scientist Intern

  • May 2021 - Nov 2021

  La Défense, Île-de-France, France During my internship, I worked on predicting train bookings and revenues. This involved considering factors like train schedules, dates, and where the train was going from and to, as well as how many days were left before the train set off. The main objectives were to help set ticket prices and to share a report on how busy the train stations across France were likely to be - this was to help plan the station staff schedules. The project involved Python coding, with some SQL, using… Show more During my internship, I worked on predicting train bookings and revenues. This involved considering factors like train schedules, dates, and where the train was going from and to, as well as how many days were left before the train set off. The main objectives were to help set ticket prices and to share a report on how busy the train stations across France were likely to be - this was to help plan the station staff schedules. The project involved Python coding, with some SQL, using notebooks and scripts in PyCharm. For the production side of things, we used tools like Jenkins and Azure, with Git for tracking changes and Poetry for handling dependencies. For the model itself, I adapted some models for time series prediction, used time series clustering to speed up queries and performance dashboards. Show less



CNRS - Centre national de la recherche scientifique

• ML in Finance Research Intern

  • Jun 2020 - Oct 2020

  Palaiseau, Île-de-France, France As a research intern I used AI techniques, specifically deep neural networks with TensorFlow, to solve a high-dimensional partial differential equation. This approach was pivotal in determining the optimal strategy for electricity sellers in renewable energy markets. In the process, I delved deeply into the Universal Approximation Theorem for neural networks, which posits that neural networks can approximate any function, including the solution to a high-dimensional partial differential… Show more As a research intern I used AI techniques, specifically deep neural networks with TensorFlow, to solve a high-dimensional partial differential equation. This approach was pivotal in determining the optimal strategy for electricity sellers in renewable energy markets. In the process, I delved deeply into the Universal Approximation Theorem for neural networks, which posits that neural networks can approximate any function, including the solution to a high-dimensional partial differential equation. This project effectively combined my interests in machine learning, finance, and renewable energy, providing innovative solutions to complex challenges faced by renewable energy providers in today's market. Show less



Banque de France

• France
• Banking
• 700 & Above Employee

• Consultant - Applied Statistics Project

  • Sep 2019 - May 2020

  As part of my second year at ENSAE Paris, I conducted an applied statistics project in partnership with the Bank of France. The objective was to identify underlying states in the economic and financial system, essentially predicting periods of economic crisis and prosperity. We employed a variety of forecasting techniques and carried out extensive feature selection to build robust models capable of anticipating shifts in the economic landscape. This project underscored the power of statistical… Show more As part of my second year at ENSAE Paris, I conducted an applied statistics project in partnership with the Bank of France. The objective was to identify underlying states in the economic and financial system, essentially predicting periods of economic crisis and prosperity. We employed a variety of forecasting techniques and carried out extensive feature selection to build robust models capable of anticipating shifts in the economic landscape. This project underscored the power of statistical methods and machine learning in the field of economics, providing valuable insights into the dynamics of financial markets. Show less



Rothschild & Co

• France
• Financial Services
• 700 & Above Employee

• Quantitative Analyst Intern

  • Jun 2019 - Aug 2019

  Monaco As a Quantitative Analyst intern in portfolio management, I developed a three-criteria quantitative model that generated trading signals (strong sell, sell, neutral, buy, strong buy) for risky assets, primarily stocks from various companies. The model, based on a deterministic decision tree, required careful selection of criteria, leveraging previous research to accurately determine buy or sell signals for equity portfolios. Backtested over the period 2000-2019, the model significantly… Show more As a Quantitative Analyst intern in portfolio management, I developed a three-criteria quantitative model that generated trading signals (strong sell, sell, neutral, buy, strong buy) for risky assets, primarily stocks from various companies. The model, based on a deterministic decision tree, required careful selection of criteria, leveraging previous research to accurately determine buy or sell signals for equity portfolios. Backtested over the period 2000-2019, the model significantly outperformed the benchmark, demonstrating resilience even during periods of economic downturns. Show less



KRYPTOSPHERE®

• France
• Higher Education
• 1 - 100 Employee

• Head Of Treasury

  • Oct 2018 - Jun 2019

  Île-de-France Treasurer at KryptoSphère Association. Our mission is to demystify Blockchain technology for a broad audience, primarily through organizing conferences. The unique aspect of KryptoSphère ENSAE is our focus on developing computer techniques and writing code in Solidity.



Festival de Ramatuelle

• France
• Performing Arts
• 1 - 100 Employee

• Usher/Controller

  • Jul 2012 - Sep 2018

  Ramatuelle For seven consecutive summers, I held a student job at the Ramatuelle Festival. Initially starting as an usher, I was responsible for seating guests, and later advanced to the position of controller, where I managed attendee access. This experience exposed me to a wealth of culture and introduced me to various artists, including actors, magicians, and singers. The exposure to such a diverse range of performances and the cultural richness it brought has been a truly enriching experience.



Centre des Impôts

• France
• Government Administration
• 1 - 100 Employee

• Employé

  • Jun 2012 - Jul 2012

  Saint-Tropez, Provence-Alpes-Côte d’Azur, France Summer job at the Tax Center. Gained valuable experience with software tools used for identification and classification of tax declarations. This role offered me a practical insight into the operation of public finance institutions and the application of IT systems in a real-world context.