I grew up in the south of France and graduated from the Mines Saint-Etienne, one of the top engineering schools in France. During the first years of my engineering school, I enjoyed a lot learning about mathematics and I decided to pass exams to graduate with a mathematical bachelor at the Université Jean Monnet - Saint-Etienne.
To learn even more about mathematics, I spent a semester at the University of Kaiserslautern (Germany) following courses about statistics, integer programming, and partial differential equations. This semester was a really great experience because I met a lot of different people from different cultures and I even had the chance to play in a rugby team composed mainly of US soldiers.
When I came back to France, I started following courses about big data and statistics. I then discovered data science and machine learning and decided to do a master in science in data science, partial differential equations, and stochastic calculus.
Right before my master, I did a 3-month internship at the I3S laboratory about statistics of particle's cellular automata. During this internship, I helped a professor write an article and provide proofs for several lemmas and propositions.
When I started to learn about data science and machine learning, our professors encouraged us to participate in a challenge created by Total. The goal of this competition was to predict as accurately as possible, the sales of several fresh products (sandwiches, drinks, etc.) in two gas stations. After making one of the top predictions, and writing a report for Total, I won the challenge.
The reward for the challenge was a 6-month internship of my choice for Total, so I decided to work on real-time anomaly detection during oil well drilling. I learned a lot about machine learning as I was working in a team of six data scientists and one drilling engineer. In particular, we managed to create models to predict two of the most recurrent problems occurring during oil well drilling (stuck pipe and kick).
After graduating from my engineering school, I decided to create my own websites and tools online, mostly in ruby on rails and python. The first project I did was UniquePDF, a secured file sharing tool that uses unique watermarks. This tool is still up and running and we are serving several companies. I wrote several blog articles in order to increase our ranking on Google and get more organic traffic.
The second SaaS I created is ChurnTarget, a mix of a tool and service I provide for SaaS that wants to reduce their churn. I created an integration on Intercom to make it easier for me to gather the company's data about their clients. After gathering and analyzing their data, I create a custom machine learning model to predict which customers are at risk of churning in the following weeks or months. These predictions can help their customer support teams to focus their efforts on risky customers.
Although the objective is always to reduce their churn, working for different companies is a different experience each time as their main activity and clients are completely different.
I love playing a lot of sports and listening to music. Since I'm seven years old, I have been playing rugby and playing guitar. I also like to hang out with my friends and hike occasionally.