Freelancer profile translated to English.
Back to original languageAbout Nada
Graduated from an engineering school and with a master's degree in artificial intelligence, I have 6 years of experience in data. I have had the opportunity to work on several Data Science problems, combining different types of data: texts, images, structured data, and time series.
Excellent command of Machine learning and Deep Learning algorithms, and associated python libraries such as: Pandas, Numpy, Scikit-learn, Tensorflow, Pythorch, HuggingFace.
I have also worked on MLOps projects, from the experimentation phase to production deployment on the cloud, setting up the entire ML pipeline (ETL creation, ML model research and training, task sequencing with kedro, auto tests, docker, github, MLFlow, Rest API, docker, CICD Pipeline).
French
Native or bilingual
English
Fluent
Arabic
Native or bilingual
Spanish
Conversational
Can work on-site
Paris (up to 50km)
Experience
- TF1Lead Data Scientist / MLOpsPRESS AND MEDIAJanuary 2025 - Today (1 year and 7 months)Boulogne-Billancourt, France
- Management of Machine Learning projects from POC to production deployment
- Building and training Machine Learning and Deep Learning models on structured and unstructured data for various applications such as: Recommendation on MyTF1 and TF1Info, SEO optimization for the TF1Info website, Predictive analysis for MyTF1 advertising targeting...
- In-depth experience with Snowflake for querying and analyzing Big Data
- Data preprocessing and ML modeling on Databricks with tracking via Mlflow.
- Design and planning of Airflow DAGs to automate model training pipelines.
- Development of APIs to expose models, use of Azure DevOps for code versioning, and CI/CD pipelines including security and load tests.
- Deployment of containers in Kubernetes clusters to optimize application performance and scalability.
- Real-time monitoring of models in production, particularly algorithm response times, success rates, and request counts, accompanied by alerting systems.
- Brut.Lead Data ScientistFILM AND AVOctober 2021 - December 2021 (3 months)Paris, FranceData Lead for Brutx: the company's streaming platform Brut.- Audit of existing data quality and solicitation of partners for new relevant data necessary for our studies.- Data mining on large volumes of data to calculate company KPIs (Recruits, subscribers, churns, revenues, etc.).- Customer segmentation analysis- ML model for predicting churn rate based on acquisition channel, platform consumption quantity, seasonality, etc...NLP model on churner questionnaire data (sentiment analysis, topic modeling...) for improving Brutx content and reducing churn rate.NLP model to predict the open rate of newsletter emails.
- Sorbonne UniversitéResearch in Natural Language ProcessingRESEARCHJune 2021 - Today (5 years and 2 months)Paris, FranceAs part of a research project in Data to Text at Sorbonne University, the challenge was to create a Deep Learning model capable of generating textual data from structured data and vice versa.Application: generation of natural language summaries from time series / structured data, or extraction of key figures from textual data.- Study of the state-of-the-art in NLP and Data to Text. Choice of the pre-trained Transformer T5 model and adaptation of this model to the Data to Text and Text to Data task- Supervised training (Transfer Learning with HuggingFace)- Unsupervised cyclic training of the T5 model (Transfer Learning with HuggingFace)- Modification of the T5 model architecture to integrate a probabilistic latent space (Conditional Variational Autoencoders) to control the generation of text and structured data- Distributed training on multiple GPUs. Tracking with MLFLOW
Recommendations
Be the first to recommend Nada
Help this freelancer shine by sharing your experience working together.
These freelancer profiles also match your criteria
Agatha Frydrych
Backend Java Software Engineer
4.7
(3)
2
Baptiste Duhen
Fullstack developer
4.6
(4)
5
Amed Hamou
Senior Lead Developer
4
(2)
7
Audrey Champion
Web developer
4.3
(3)
4
Education
- Applied Mathematics EngineerEcole Centrale de Nantes2015- Spécialité: Mathématiques Appliquées et Finance d'entreprise - Tronc commun: Mathématiques, Mécanique, Electronique, Physique - Cours transversaux: Management de projet, Finance, Economie, Communication...
- Master Artificial Intelligence, Systems, DataUniversité Paris Dauphine - ENS Paris - Mines PariTech2020- Modèles de Machine Learning et de Deep Learning - Traitement de données et Machine learning en environnement Big Data - Computer Vision - Natural Language Processing - Apprentissage par renforcement et Recherche Arborescente de Monte Carlo - IA on the cloud : AWS (IA services et Sagemaker) - Traitement de données en streaming avec Kafka et Flink - Données NoSQL ( Neo4, MangoDB et ArangoDB) - Data visualization (Tableau , plotly, bokeh)