Machine learning engineer based in Caen, France. I work on computer vision and LLM applications. For my master's thesis at the GREYC lab, I studied how deepfake detectors can be fooled by adversarial attacks.
I'm looking for an ML / AI Engineer position in France.
| Project | What it does | Built with |
|---|---|---|
| adversarial-attacks-rag · live demo | Answers questions on research papers about adversarial attacks, in English or French, and cites the paper and page behind every claim. | sentence-transformers, Gemini, FastAPI, Streamlit, Docker |
| network-intrusion-detection | Intrusion classifier tested on attack types it never saw in training, explained with SHAP and served as a tested API. | scikit-learn, SHAP, FastAPI, Docker |
| driver-drowsiness-detection | Real-time drowsiness monitoring from one camera, deciding on eye closure over time rather than on a single frame. | YOLOv8, MediaPipe, OpenCV |
| ai-personal-trainer | Counts repetitions and checks form from a single camera, adapting to each person's range of movement. | MediaPipe, YOLOv8, Streamlit |
Python · PyTorch · scikit-learn · OpenCV · Hugging Face · FastAPI · Streamlit · Docker · Git