Arnaud Fouchet
Verified Expert in Engineering
Data Scientist and Developer
Arnaud is a data scientist with eight years of experience both as a freelance and a team lead. He likes to communicate within teams, helping them grow, setting objectives, and establishing processes. Arnoaud turns a business need into a mathematic problem and quickly moves from an idea to POC.
Portfolio
Experience
Availability
Preferred Environment
Python, Scikit-learn, TensorFlow, PyTorch, Pytest
The most amazing...
...thing I've built is an online-ad bidder that gave us better performances than Google's.
Work Experience
Freelance Data Scientist
Pernod Ricard
- Analyzed sales information and marketing spending to identify the ROI of all marketing spend using a Bayesian model. I was in charge of improving the model for a specific country, Spain, especially considering COVID-19.
- Worked on improving the model, arriving at a specific mathematical formulation, and developing the learning algorithm from scratch using PyTorch.
- Developed a demand-forecast AI as a side project. Seeing the results of my MVP, the group decided to launch a full team of six people on the project.
CEO | CTO
Talk2me
- Created all the code, back end, iOS, and Android app.
- Founded the company, found two co-founders, and established our market positioning.
- Created the marketing campaign, and made many focus groups to validate our product and campaigns.
Lead R&D
Dolead
- Developed an online-ads bidder that was competitive with Google's. We A/B tested our bidders against Google's one and our bidder made a +35% margin in our 2019 A/B test.
- Managed a team of four people, in charge of researching algorithms, implementing them, and maintaining them. Additionally, I have experience maintaining a project that managed millions of euros monthly.
- Was in charge of communication of the R&D in the company but also to external conferences, making presentations to clients.
Experience
Online-ad bidder Dolead
https://www.youtube.com/watch?v=Ip_e3_8rQZUWe ensured that we stayed competitive by regularly doing an A/B test of our bidder against Google's. In 2017 and 2019, we made more margin using our bidder; Google's bidder got more volume, but with a higher cost per acquisition, and overall was less profitable.
To achieve this performance, we used many techniques:
• Deep learning and transfer learning; learning the deep architecture of the neural network on one task where we had more data, then fine-tuning on our pricing task, which had fewer data.
• Mixture of models; using the average of several models got better performance than using the best one. Also, instead of using the average, you can learn your optimal weighted average via machine learning.
Talk2me Dating App
http://www.talk2me.ioI built the back end and front end using Python, Django, and React Native.
We had a very experiment-driven process, so we made plenty of focus groups to validate our app, ideas, and marketing campaign.
The application was released in mid-December.
Skills
Languages
Python, SQL, Snowflake
Libraries/APIs
Scikit-learn, TensorFlow, PyTorch, React
Tools
Pytest, Jira, MATLAB, Ansible
Other
Machine Learning, Mathematics, Computer Science
Frameworks
Django
Paradigms
Distributed Computing
Platforms
Firebase, Azure, Databricks
Storage
MongoDB, Redis
Education
Ph.D. Degree in Informatics and Applied Mathematics
University of Évry Val d'Essonne - Évry, Île-de-France, France
Master Vision Apprentissage in Machine Learning
E.N.S. Cachan - Cachan, France
Master's Degree in Mathematics and Computer Science
École des Ponts ParisTech - Champs-sur-Marne, France
How to Work with Toptal
Toptal matches you directly with global industry experts from our network in hours—not weeks or months.
Share your needs
Choose your talent
Start your risk-free talent trial
Top talent is in high demand.
Start hiring