Freelancer profile translated to English.

Description

I am a Senior Data Scientist with over 7 years of experience in data science, specializing in the design of recommendation engines, natural language processing (NLP), and the implementation of MLOps solutions. I have worked with technologies such as Python, PySpark, PyTorch, and Docker in various sectors, including finance, automotive, and market research. A graduate in telecommunications engineering, I possess strong skills in data analysis, predictive modeling, and deep learning. I am also certified in Azure and Deep Learning, and I am bilingual in French and English.

Languages

French
Native or bilingual
English
Native or bilingual
Arabic
Native or bilingual

Workplace preferences

Can work on-site

Paris (up to 50km)

Société Générale
Machine Learning Engineer
BANKING AND INSURANCE
March 2021 - June 2024 (3 years and 3 months)
Paris, France
• Design and implementation of a hybrid recommendation engine (content-based, collaborative filtering, and time-based) to assist group customer advisors.
• Evaluation of model performance (3.5 times better than the historical solution) and stabilization of results.
• Implementation of an explainability module to justify recommendations made to advisors.
• Industrialization of the recommendation engine through continuous training pipelines with automatic validation.
• Conducted several behavioral studies on retail bank customers (BDDF) (post-dispute behavior, churn upon expiry of savings offers).
• Implementation of an AI assistant using LLM technology, leveraging internal bank documents via a RAG pipeline.
• Segmentation and embedding of internal documents into a vector database (Milvus) to enrich the AI assistant's context.
• Led an internal community focused on MLOps best practices and the development of reusable modules.
• Technical environment: Python, Pyspark, PyTorch, Kedro, MLflow, GitHub Actions, Snowflake, Teradata, Hive, Linux, Docker.
Infopro Digital
Data Scientist Consultant
TECH
February 2020 - March 2021 (1 year and 1 month)
Paris, France
• Geolocation of articles from a magazine (L'Argus de l'assurance) for digital marketing.
• State-of-the-art review of Named Entity Recognition (NER) technologies to detect local authorities.
• Indexing of the Geonames geographic directory on an ElasticSearch node.
• Implementation of the data access layer and query refinement for toponym resolution.
• Design and implementation of a magazine article recommendation engine based on reading history (reading sessions, fading coefficient) and article similarity (content-based).
• Deployment via a Flask API for integration and access to developed functionalities.
• Technical environment: Python, Pycharm, Spacy, Docker, Solr, ElasticSearch, Flask, AWS, Databricks, MLflow.
Ipsos
Data Scientist Consultant
September 2019 - February 2020 (5 months)
Paris, France
• Participation in the implementation of an advertising panel evaluation solution based on traffic data.
• Collection, structuring, and construction of various databases (customers/POI/travel).
• Project structuring (code modularization) and integration of different codebase components.
• Pre-processing and exploratory data analysis (EDA) of multimodal traffic flows.
• Modeling and prediction of French travel vectors (Speed, Distance, Times, modes, etc.) using a multi-task learning model.
• Technical environment: Pycharm, Python, PgSQL, numpy toolkit, gitlab, Git, Jupyter, Docker, tensorflow, shapely.