Machine Learning Engineer @ LVMH · Founder @ SetAIComply
Production AI for luxury Maisons — GenAI & LLMs · Recommender Systems · Computer Vision & NLP · MLOps
I design and ship production machine learning and generative AI systems — from retrieval and recommendation to LLM assistants — and I care as much about what happens after the model is trained as before.
- 🏛️ ML Engineer at LVMH on the ATOM AI platform, building ML & GenAI products for the Group's Maisons (clienteling, recommendation, forecasting, marketing science, computer vision, NLP and Responsible AI), delivered as reusable Dataiku DSS plugins and GCP micro-services.
- 🛡️ Founder of SetAIComply — AI-Act-native compliance software for European SMEs: classify AI systems, generate Annex IV technical documentation and stay compliant in 24 EU languages.
- 🎓 M2 Data Science, Université Paris-Saclay · Statistical Engineer, ENSAE Pierre Ndiaye · previously IBM, Carrefour, FeetMe.
- 🇪🇺 Currently deep into AI governance & the EU AI Act, and how to make compliance actually usable by small teams.
| Area | Focus |
|---|---|
| GenAI & LLMs | RAG assistants, agentic workflows, LangChain / LangGraph / llama-index, evaluation, reverse-explainability & Responsible AI |
| Recommender systems | Two-Tower retrieval, Deep & Cross Networks for cold-start, sequential models with Direct Preference Optimization, embeddings |
| Computer vision & NLP | CLIP & NASNet image embeddings, OpenCV pipelines, multilingual sentence-transformers, BERTopic, KeyBERT, UMAP |
| Marketing science | Deep-learning attribution, Bayesian & OLS Marketing Mix Modeling |
| MLOps | Vertex AI, Dataiku DSS, MLflow, Optuna, Docker, FastAPI, OpenTelemetry, GitHub Actions CI/CD |
| AI governance | EU AI Act conformity, risk classification, Annex IV documentation, evidence automation |
Languages
ML & Deep Learning
GenAI
Cloud & MLOps
| Project | What it is | Status |
|---|---|---|
| SetAIComply | EU AI Act compliance SaaS for European SMEs — system classification, Annex IV documentation, continuous evidence, 24 EU languages. FastAPI + Next.js. | Live |
| CitezMoi | AI-visibility audit for French-speaking SMEs and agencies: measure how you are cited by ChatGPT & co., get generated fixes and weekly tracking. | In development |
| ATOM AI @ LVMH | Production ML & GenAI products for the Group's Maisons, packaged as reusable Dataiku DSS plugins and GCP micro-services. | Ongoing |
| Repository | What you will find | |
|---|---|---|
| Computer-Vision-for-Autonomous-Cars | How self-driving perception works: lane detection, object detection & tracking, LiDAR — with hands-on notebooks |  |
| Reinforcement-Learning-Specialization | Full walk-through of the University of Alberta Fundamentals of Reinforcement Learning specialization |  |
| GCP-Professional-Data-Engineer-Learning-Path | The path I followed to get Google Cloud Professional Data Engineer certified |  |
| Human-Activity-Recognition | Activity recognition from wearable inertial sensor networks, from feature engineering to validation protocol |  |
| Virtual-Adversarial-Training | Keras implementation of the original VAT paper for semi-supervised learning |  |
| Data-Science-With-Python | A practical introduction to data science with Python — pandas, NumPy, matplotlib |  |
➜ 60+ public repositories on my repositories page, covering ML, deep learning, GCP, big data and statistics.
I publish notes and tutorials on data science and AI on Medium:
- M2 Mathematics & Applications — Data Science, Université Paris-Saclay (co-accredited with CentraleSupélec, École Polytechnique, ENS Paris-Saclay, ENSAE Paris, Télécom SudParis, ENSIIE)
- Statistical Engineer, École Nationale de la Statistique et de l'Analyse économique Pierre Ndiaye
- 19 certifications, including Google Cloud Professional Data Engineer, Machine Learning (Duke University), NLP with Classification and Vector Spaces (DeepLearning.AI) and the IBM Enterprise Design Thinking track.
I'm always happy to discuss applied GenAI in production, recommender systems at scale, or making the EU AI Act workable for small teams.
