Antoine Lucas
I trained as an agronomy engineer, specialized in applied statistics, and somehow ended up building machine learning systems for Chanel’s fragrance R&D — and fine-tuning LLMs for scientific and regulated contexts. The path made sense at each step, even if it doesn’t look obvious on paper.
For the past 5+ years I’ve been embedded in R&D and clinical teams at L’Oréal, Sanofi, Abolis (microbiome biotech), and now Chanel Parfums Beauté — building ML systems at the intersection of data and experimental science.
Currently Data Scientist & ML Engineer at Chanel Parfums Beauté R&D, on mission via Astek / IT&M Stats. Building ML and computer vision tools for fragrance and cosmetics research.
What I do
Machine learning & deep learning
Computer vision models, NLP pipelines, and LLM fine-tuning for scientific applications. From texture classification and olfactory descriptor mining to training LoRA adapters on domain-specific datasets. Production-grade tools that go into daily research workflows.
LLM pipelines & agentic systems
End-to-end pipelines combining ASR, structured extraction (Instructor + Pydantic), and domain validation — turning unstructured scientific inputs into validated, auditable records. Local-first, compliance-aware architecture.
Statistical modeling & analysis
From design of experiments and mixed models to biostatistics in Good Practices (GxP) environments. I cover the full analytical chain — from raw assay data to interpretation for regulatory submissions.
Reproducible data science & MLOps
Everything in Git, containerized, runnable in 6 months with uv sync or renv::restore(). Drift monitoring, CI/CD pipelines, and model versioning in GxP-adjacent environments. I’m opinionated about this because I’ve seen what happens when you’re not.
Skills
Languages
Python R SQL Bash
Deep learning & LLM
PyTorch Hugging Face PEFT / LoRA Instructor faster-whisper sentence-transformers
ML & Python ecosystem
scikit-learn OpenCV Pandas FastAPI uv
Statistics & methods
DoE Mixed Models Bayesian Optimisation PLS Survival Analysis SPC
R ecosystem
Shiny Tidyverse ggplot2 renv Quarto
MLOps & infrastructure
Docker GitHub Actions Azure ML Databricks Posit Connect GxP / Regulated
Education
- Diplôme d’Ingénieur — Statistiques appliquées aux Sciences de la Vie — Institut Agro - Agrocampus Ouest, Rennes · 2021
- Master — Mathématiques Appliquées · Statistiques — INSA Rennes · Agrocampus Ouest · Université Rennes 2 · ENSAI · ENSAE (joint programme) · 2021