Exploring the intersection of Chemistry, Data Science, and AI models.
I design open-source tools that combine scientific rigor and digital innovation. From molecular discovery to AI-powered assistants, my goal is to help scientists push the boundaries of their research.
I am independant, and always open to collaborations.
- Deep learning for chemistry and molecular discovery
- Photochemistry, photocatalysis and light-driven processes
- Biostimulants and sustainable agriculture through data-driven insights
- Intelligent assistants and productivity tools for science (RAG, agentic workflows, and LLM fine-tuning)
-
Bioactivity prediction using Graph Neural Networks on molecular structures. Trained on ChEMBL data, this model predicts whether a compound is active against specific biological targets.
👉 See the presentation of the project (in French!) -
Prediction of photochemical properties (HOMO-LUMO gap, orbital energies) using Graph Neural Networks. Leverages quantum chemistry datasets (QM9, PC9, Transition1x, Harvard OPV) to train models capable of accelerating photocatalyst design and organic solar cell discovery.
-
A modular chat interface to interact with local models that maintain persistent memory, based on Local LLM Memorization with automatic update and retrieval.
-
Enhance interactions with local models by automatically memorizing and summarizing past conversations to generate precise, context-aware prompts. Built around a lightweight local SQLite database for easy memory access.
-
An automated bibliographic-database system that integrates scientific papers into a searchable graph for advanced literature analysis. Features an AI chatbot that can query and "discuss" directly with the user's bibliography.
If you're working on projects related to sustainable chemistry, clean energy, or molecular innovation, feel free to reach out — I'd be happy to discuss with like-minded scientists!
