Engineering student in artificial intelligence at ENSTA Paris (Institut Polytechnique de Paris). Looking for a final-year internship (PFE, from April 2027) in AI R&D or MLOps, in the Aix-Marseille area.
My work sits between research and engineering: training and evaluating deep-learning models, and building the pipeline around them — datasets, evaluation protocol, CI, containers, self-hosted infrastructure.
| Paper | Venue |
|---|---|
| Player-Specific Policy Adaptation in Elite Chess with MoE-LoRA and Limited Search | BNAIC/BeNeLearn 2026 — oral |
| Toward Modeling Player-Specific Chess Behaviors | arXiv:2605.11893 (cs.AI) |
| Repository | Content | Stack |
|---|---|---|
| ChessBehaviors | Research code: PGN dataset build, Maia-2 variants (fine-tuning, nucleus pruning, MCTS, MoE-LoRA), JSD and style-space evaluation, LaTeX tables | Python, PyTorch |
| OxonoProject | Agents for the Oxono board game: minimax, MCTS with PUCT, AlphaZero, Cython training loop | Python, Cython, PyTorch |
| TIPE-Jeux-De-Dames | Draughts engine (10×10 rules) with a Pygame interface and a neuroevolution-trained agent | Python, PyTorch |
| PP.GG | League of Legends custom-game MMR tracker: desktop client, Flask API, Discord bot, Docker | Flask, Discord, Docker |
| IN204_projet | Doom-style 2.5D raycasting engine written from scratch | C++, SDL2 |
| Thermostat-Radiateur | Raspberry Pi thermostat driving a 433 MHz RF heater | Python, GPIO |
Python · PyTorch · pandas / NumPy · C / C++ · Docker · Git and CI · Linux · Proxmox VE / Ceph

