Hey, I’m Cédric 👋 — currently building foundation models and multimodal systems at Lunit to help rewire how we diagnose and treat cancer.
My day-to-day is a mix of deep computer vision, large-scale training, and occasionally convincing GPUs to behave.
- Foundation Models for Oncology — training large-scale vision models on digital histopathology to find new biomarkers and improve diagnosis.
- Multimodal AI for Cancer Research — integrating histopathology with other modalities (the fun ones are still under wraps 🤫) for a full-spectrum view of cancer.
- Computer Vision at Scale — from microscopes to gigapixel slides, making sense of the tiny details that matter.
To make AI that’s robust, interpretable, and useful enough to actually change clinical practice — starting with cancer.
If it needs to reason across images, text, chemistry, and biology, I’m probably interested.
| Project | What It’s About |
|---|---|
| Foundation Model for Cell Painting | Large-scale vision model learning rich, generalizable features from cell morphology images. |
| Multimodal Phenotype–Compound Translation | Bidirectional model translating between cell painting phenotypes and compound SMILES. |
| Educatif PanNuke Segmentation | Educational CV project on semantic segmentation, implementing U-Net & U-Net++ from scratch in PyTorch. |
| Neural Style Transfer | Reimplementation of the classic algorithm — because sometimes you do just want Van Gogh to paint your cat. |
Best place to find me is on LinkedIn.