Skip to content
View CedricCaruzzo's full-sized avatar

Block or report CedricCaruzzo

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
CedricCaruzzo/README.md

LinkedIn

Hey, I’m Cédric 👋

I am an AI Research Scientist at Lunit, bridging the gap between Academic Research and Industrial-Scale Engineering.

I build AI that doesn't just "see" medical images but reasons about them. My work focuses on Foundation Models, Agentic AI, and the Distributed Infrastructure required to train them on Petabyte-scale biomedical data.


🧠 Current Focus

  • Agentic Clinical Intelligence — Moving beyond standard SFT to build multi-agent systems that mine clinical reasoning traces from raw data.
  • Foundation Models (ViT-H / DINOv2 / DINOv3) — Adapting Self-Supervised Learning to high-resolution digital histopathology to find robust biomarkers.
  • Distributed Systems at Scale — Architecting fault-tolerant orchestration pipelines to process hundreds of Terabytes of WSI data without breaking a sweat.

🚀 Selected Public Projects

While my current work on Foundation Models is proprietary, here is some of my open research:

Project What It’s About
Foundation Model for Cell Painting Large-scale vision model learning rich, generalizable features from cell morphology images (KAIST Master's Thesis).
Multimodal Phenotype–Compound Translation Bidirectional model translating between cell painting phenotypes and compound SMILES.
PanNuke Segmentation (Educational) Implementing U-Net & U-Net++ from scratch in PyTorch for semantic segmentation on histopathology data.
Neural Style Transfer A clean reimplementation of the classic algorithm — because sometimes you just want Van Gogh to paint your cat.

🎯 My Mission

To build AI that is robust, interpretable, and capable of end-to-end clinical reasoning. I am interested in systems that connect the dots between pixels, biology, and text.


📬 Reach Me

The best place to find me (and see my latest updates) is on LinkedIn.

Pinned Loading

  1. Racial-Bias-Removal-for-Face-Age-Progression Racial-Bias-Removal-for-Face-Age-Progression Public

    This project introduces a method to mitigate demographic bias in generative face aging models. We adapt the StyleGAN-based SAM model by adding a race-consistency loss, enforced by the DeepFace clas…

    Python

  2. pannuke-segmentation pannuke-segmentation Public

    An open-source UNet-based pipeline for nuclei segmentation in histopathology images using the PanNuke dataset. It features an interactive web app for easy data visualization and handling, making AI…

    Python 9 5

  3. Neural-Style-Transfer Neural-Style-Transfer Public

    Implementation of Neural Style Transfer based on the original paper by Gatys et al., with adaptations from Deep Learning with Python by François Chollet. This notebook allows you to experiment with…

    Jupyter Notebook 4 2

  4. minimal-word-embeddings minimal-word-embeddings Public

    An educational project building Word2Vec from scratch in PyTorch to demystify how machines can learn semantic meaning from raw text.

    Python