Description

I worked as a data scientist at Reezocar. I developed skills to lead projects from ideas to production. Data analysis, machine learning and when necessary deep learning. The last few years I moved onto data engineering and data architecture in general.

Recently I've had an interest in the AI capacity to generalize the understanding of a problem and the ARC (Abstraction and Reasoning Corpus) challenge specifically.

Langues

Français
Bilingue ou natif
Anglais
Capacité professionnelle complète
Espagnol
Notions

Préférences en matière de lieu de travail

En télétravail uniquement

Travaille majoritairement à distance

SAS Check&Visit
R&D / Data Engineer
IMMOBILIER
septembre 2020 - Aujourd'hui (5 ans et 9 mois)
Data Engineer

Machine Learning - Deep Learning - Mobile integration of DL models - REST API - Serverless - Data Architecture
Machine learning Deep Learning API REST FastAPI GraphQL Serverless Docker docker-compose
Reezocar
Data Scientist
AUTOMOBILE
juillet 2017 - juillet 2020 (3 ans)
Paris, France
Machine Learning - Deep Learning - Image Processing - Natural Language Processing - Data Analysis - Recommender System - Lambda Architecture - Azure Datalake

The mission of Reezocar is to facilitate the research and to secure the purchase of second-hand cars in Europe. To that end, Reezocar gathers millions of second-hand cars ads.

The Data Science team has multiple missions :
- Ensure data quality and availability
- Design solutions to business issues based on data modelling
- Develop, deploy and maintain these solution

Examples of projects deployed include :
- NLP matching algorithm (based on Microsoft publication) to extract information from unstructured products.
- Recommender system to match the best products based on a client project
- Prediction of market value variation
- Image processing for classification and filtering
- Design and implementation of an Hadoop-based Datalake with 500M+ documents to meet the Data Science Team needs

Technologies : Python, Numpy, Pandas, Scikit, Graphql, Hadoop, Hive, Spark, Data Factory, ElasticSearch, SQL, MongoDB, Redis, Azure, Jenkins, Git, Docker, Kubernetes, etc
Python Kubernetes GraphQL Jenkins Microsoft Azure MongoDB Redis Docker Elasticsearch Scikit-learn Pandas Numpy Hadoop Hive MySQL
Pumpkin App
Data Scientist
BANQUE & ASSURANCES
septembre 2016 - juin 2017 (9 mois)
Paris, France
The FinTech world is on the edge of revolution. These fast growing companies aim at a better quality of service along with a more stable financial landscape. Built on the merging of financial services and communications technology, these companies go from payments to wealth management, crowdfunding to peer-to-peer lending.

Pumpkin is one of the main actors in the french FinTech with fast growing community. Pumpkin app is a simple and secure mean of payment among friends. Pumpkin is drove by the conviction that paying a friend should be as simple as sending a text. I was in charge of developing a deeper knowledge of the database to improve decision making.

• Multi-scale reports
• Data dashboards & templates for specialized team
• Activation & retention strategy
• Automate data refresh & standalone reports
• Mini-Websites development

--- Technical skills ---
• DAX, powerBI (Microsoft)
• Python Scientific & Analytic packages (Pandas, NumPy)
• Web development (HTML, CSS, Javascript, PHP)
• Matlab

--- Human skills ---
• Efficiency & Productivity
• Synthetic thinking
Python Microsoft Power BI HTML5 JavaScript PHP