Freelancer profile translated to English.

Description

Data Scientist for 5 years, I have developed expertise in text processing (Natural Language Processing) through various assignments (finance, marketing...).

My first experience at Ekimetrics placed me at the intersection of business and technical issues. Initially as a developer on NLP (transformers, nltk) and big data (polars, spark) assignments. I was able to start there on GenAI topics.

I also have several experiences in MMM and quantitative marketing.

Finally, in my last role, I developed real API/NLP product expertise related to LLMs and RAG. Particularly with tools like LangChain, QDrant, FAISS, and various LLMs (Mistral, OpenAI, LLaMA). I was also tasked with fine-tuning an LLM model for the medical sector. Approaching AI from a medical angle allowed me to understand the challenges of data, quality, and reliability of solutions in a different way.

Today, I work on NLP, MMM, and other Machine Learning assignments for various industries.

Key elements:

NLP for finance:

- Creation of NLP scoring pipelines for financial markets

- Creation and industrialization of investment strategies

- Creation of a RAG tool on financial articles with OpenAI

Performance Marketing:

- Marketing Mix Modeling for automotive, luxury, retail, and gaming industries

E-consent platform:

- Drafting of specifications and database schema

- Testing and feedback with a developer and an UX/UI designer

- Continuous development of the app

On a personal note, I maintain a strong interest in Machine Learning, particularly Deep Learning, and am an active member of the HuggingFace community.

I am available to assist you with data analysis assignments, NLP POCs, or production deployment with your technical team.

Looking forward to connecting with you!

Industry field of expertise

Languages

French
Native or bilingual
English
Fluent

Workplace preferences

Can work on-site

Paris (up to 50km), Lyon (up to 30km), Nantes (up to 30km), Marseille (up to 10km), Lille (up to 15km)

L'Oréal - Innovation
NLP Data Scientist
FASHION AND COSMETICS
September 2024 - March 2026 (1 year and 5 months)
Paris, France
NLP applied to the detection of topics mentioned in customer reviews and the analysis of their sentiments.
Design, evaluation, and deployment of topic detection models for 7 languages, over 250 topics, and a weekly volume of 200k inputs.
Environment: Vertex AI, Kubeflow, Docker, Big Query...
Natural Language Processing (NLP) Big Query Hugging Face Topic Modeling Pytorch
NUVOCARE
Chief Technical and Product Officer
MEDICAL
September 2023 - October 2024 (1 year and 1 month)
Paris, France
Nuvocare is a startup aiming to digitize, simplify, and standardize informed patient consent before a medical procedure.

Technical tasks:
- Creation of technical specifications
- API and back-end development
- Monitoring of front-end development (tests, checks, and audits)
- Creation of Nuvochat: an LLM dedicated to health-related questions
- Guarantee of the relationship with the cloud provider

Commercial tasks:
- Prospecting by email and calls
- Client meetings and technical presentations
Product Development MongoDB OVHCloud Natural Language Processing (NLP) e-health
Ekimetrics
Senior Data Scientist
CONSULTING AND AUDITS
April 2021 - September 2023 (2 years and 5 months)
Paris, France
NLP for asset management:
- Maintain and improve a Machine Learning pipeline (Kedro) on Databricks
- Development of a thematic sentiment analysis model in TensorFlow (HuggingFace)
- Implementation of new features and management of ticket prioritization
- Regular updates with asset managers and investment professionals

Quantitative Marketing:
- Marketing mix modeling and optimization for automotive, insurance, retail, etc.
- Customer satisfaction modeling
- Development of internal methodologies for analyzing correlated variables (PCA, PLSR, LDA…)
- Analysis of content effectiveness, pricing, and promotional effects
- Geo-Marketing in the telecommunications industry

Research and Development:
- Topic modeling
- Few Shot Learning with large language models (BERT, RoBerta, GPT-3…)
Python Pytorch Hugging Face Natural Language Processing (NLP) MMM