DG

Hey! I'm Daan,

I figure out what to build, and then build it.
here's a bit of my work.

About

Daan Geijs

I'm the Product Discovery Lead at kaiko.ai, where I work out how AI and careful design can solve real problems across healthcare, and turn that into something a clinician can actually use. It's a new design space: the problems are harder than most software, so how a tool works and is understood matters as much as the model behind it.

I hold a PhD in computational pathology from Radboudumc and a double master's in biomedical engineering and physics education. Ten years of shipping models taught me that shipping is the easy part. Knowing what to ship is the hard part.

On the side I run Alpacus, my freelance practice for engineering and early product consulting. This site holds my work and writing. Reach out via work with me.

Some stuff I have done

Still writing these up, more on the way.

From the kitchen

not everything I ship is software. latest from dekeukenvandael.nl, my personal recipe book.

Writing




In the media

Publications

  • The DRAGON benchmark for clinical NLP(2025)

    Joeran Bosma et al., NPJ Digital Medicine

  • Weakly Supervised Classification of Mohs Surgical Sections Using Artificial Intelligence(2025)

    Daan Geijs et al., Modern Pathology

  • Generating dermatopathology reports from gigapixel whole slide images with HistoGPT(2025)

    Manuel Tran et al., Nature Communications

all 22 publications