$ cat ~/about.md

Gigapixel images. Natural language. Open code.

Sander Moonemans Sander Moonemans

I'm Sander, a PhD candidate at Radboudumc, building open-source vision-language models that help pathologists analyze whole-slide images and converse about them in natural language. I'm also researching multi-agent systems and foundation models for pathology.

Big believer in tea and sports, when I'm not stuck in a terminal.

location Nijmegen, NL
/
focus computational pathology
/
role PhD candidate
// interactive · type 'help' to start
sander@radboudumc: ~/research
____                  _
/ ___|  __ _ _ __   __| | ___ _ __
\___ \ / _` | '_ \ / _` |/ _ \ '__|
 ___) | (_| | | | | (_| |  __/ |
|____/ \__,_|_| |_|\__,_|\___|_|

Welcome! Type help to see available commands.

sander@web ~ $

Research & Publications

My research focuses on making AI assistants for pathologists: open, reproducible, and clinically useful. I develop vision-language models that can understand gigapixel whole-slide images and converse about them in natural language. I'm also researching multi-agent systems and foundation models for pathology.

Projects

All code, data, and models are publicly available, because medical AI research should be open.

A local-first project management tool written in Go. Combines human CLI/TUI interfaces with AI agent access through 18 purpose-built MCP tools. Your thoughts stay on your machine. Compatible with Obsidian and other markdown editors.

GoMCPCLIproductivityMIT
8

Polysome

A domain-agnostic framework for generating synthetic instruction-tuning data at scale using LLMs. Supports DAG-based prompt chaining, multiple inference engines (vLLM, llama.cpp, HuggingFace), and distributed GPU processing.

data-generationLLMPythonopen-source
6

Full-stack demo for the Kaggle MedGemma Impact Challenge based on ANTONI-Alpha. A FastAPI + Next.js application for interactive gigapixel pathology analysis with integrated whole-slide image viewer.

FastAPINext.jsdemoKaggle
1

Vision-language models for melanoma histopathology. The code from my Master thesis project, which has been published at MICCAI 2025. Generates human-level pathology reports on common lesions. Trained on 42,512 whole-slide images paired with 19,645 pathology reports.

melanomaVLMreport-generationMICCAI 2025
3 0

Curriculum Vitae

> experience

February 2025 – present

PhD Candidate: AI in Computational Pathology

Radboudumc · Nijmegen
  • Researching Large Vision-Language models for whole-slide image analysis
  • Building open-source tools for pathology AI (ANTONI-Alpha, Polysome)
  • Funded by the 2024 Ammodo Science Award
May 2023 – Feb 2025

Software Engineer (GenAI/HALO team)

CM.com · Breda
  • Built RAG-powered chatbots serving hundreds of businesses worldwide
  • Deployed transformer models in Kubernetes using NVIDIA Triton
  • Designed CI/CD pipelines via Azure DevOps, GitHub Actions, FluxCD
  • (16+ hours/week)
Sep 2022 – May 2023

Medior Backend Software Engineer

Building Blocks (CM.com) · Tilburg
  • Backend solutions using Java-Spring, deployments through Azure
Sep 2019 – Jul 2022

Development Engineer

ABN AMRO N.V. · Amsterdam
  • Backend development with Java, Spring Boot, Apache Kafka
  • Infrastructure deployments using Azure, Kubernetes (AKS)
  • Stakeholder management with business developers

> education

2025 – 2029

PhD in AI & Computational Pathology

Radboudumc · Nijmegen
Sep 2022 – Feb 2025

MSc Data Science & Artificial Intelligence

Eindhoven University of Technology · Eindhoven
Thesis: 9/10 · Average: 8.6/10
2019

BSc Artificial Intelligence

Radboud University · Nijmegen

> skills

Languages

PythonJavaSQLC++Go

AI / ML

PyTorchInstruction TuningHuggingFaceTriton KernelsNVIDIA TritonTensorRTvLLMllama.cppQuantized Models

Agents

MCP ServersAgent SkillsLangGraphClaude CodeOpenCodeGemini CLI

Cloud & DevOps

AzureGCPDockerKubernetesTerraformCI/CDAirflowNetworking

Data

Vector DatabasesOracleSQL ServerWeb Scraping

Get in Touch

contact.sh
# Feel free to reach out for collaborations,
# research discussions, or just to say hi.

export GITHUB = "github.com/SanderMoon"
export HUGGINGFACE = "huggingface.co/SaltySander"