Freelancer profile translated to English.

Description

Hello 👋

I am a data scientist with 4 years of experience, holding a dual master's degree in mathematics and computer science. My expertise lies in machine learning, data engineering, and MLOps, particularly in the cloud.

My achievements include:

Designing and deploying cutting-edge machine learning models in Cloud environments.

Optimizing pipelines (performance/costs) for data processing.

Implementing innovations and automations in artificial intelligence applications, with a focus on real-world applications.

I am passionate about innovation and technological challenges, and I am fully committed to each project to ensure superior quality results.

My education and certifications (GCP Data Engineer and AWS Machine Learning Specialty) attest to my technical skills.

Languages

French
Native or bilingual
English
Fluent

Workplace preferences

Can work on-site

Paris (up to 50km)

LexisNexis
Data Scientist - NLP Consultant
December 2021 - April 2024 (2 years and 4 months)
Paris, France
Data Science mission :
- Gathered, cleaned, and analyzed data essential for Natural Language Processing (NLP) projects, ensuring the quality and relevance of the datasets utilized.
- Designed and developed a Question-Answering model based on Retrieval Augmented Generation (RAG) using proprietary data, significantly enhancing the accuracy of responses generated by Language Model (LM).
- Implemented a cross encoder that significantly boosted the performance of RAG.
- Developed cutting-edge NLP models for document classification.
- Created NER (Named Entity Recognition), clustering, topic modeling, and embeddings models.
- Designed and implemented NLP models to predict similarity scores between documents.
- Collaborated closely with the Product team to devise and implement solutions catering to specific business needs. MLops mission :
- Led the transition from an on-premise codebase to integration within an AWS-hosted MLOps architecture leveraging MLflow, ZenML, Docker, Kubernetes, and Jenkins.
- Established automated training pipelines and batch prediction pipelines using Sagemaker, Batch Transform, and Endpoint.
- Contributed to deployment process optimization through microservices oriented architecture utilizing FastAPI and Streamlit, along with comprehensive automation.
- Optimized costs by selecting and utilizing appropriate machine types for each workflow, enabling efficient cloud resource utilization while maintaining optimal performance.
- Conducted advanced fine-tuning of Language Models on the Sagemaker platform, harnessing multi-GPU computing power to optimize model performance.
StarCLay
Data Scientist
November 2021 - April 2024 (2 years and 5 months)
Paris, France
Sports Data Streaks
Data Engineer
April 2021 - October 2021 (6 months)
Strasbourg, France

- Gathering data through various APIs.
- Designing a data storage architecture on the Google Cloud Platform (GCP).
- Developing ETL pipelines to extract, transform, and load real-time data from multiple sources using tools such as Pub/Sub, Cloud Composer, Cloud Functions, and Airflow.
- Implementing parallel statistical computations for data analysis purposes.