Clinical AI and airway systems, built at the bedside.
I turn clinical constraints into documentation systems, airway tools, and products that hold up after launch.
Selected work
Three systems, each judged by what happens when real people use them.
Sayvant
Production documentation AI designed around chart truth, physician review, and the failures that appear after launch.
100+ sites / 1.1M+ charts
IntuBlade
Single-use video laryngoscopy built around field access, lens clearing, and operational reality.
400+ agencies / 41 states
Jits Analytics
A published BJJ training journal that turns sessions, technique notes, and video into a usable progression record.
Web + iOS / publishedLatest writing
Notes on clinical AI, reliability, emergency medicine, and device work.
-
Preventing AI Fabrication in Clinical Documentation
One invented medication is enough. Clinical documentation AI has to be built around evidence.
-
Building a Chest Pain Risk Stratifier from Clinical Conversations
Chest pain exposes overconfident clinical AI. Useful support beats false diagnosis.
-
Changing Models in Clinical Documentation Without Losing Control
A new model is not a reason to touch the chart. Failure modes decide whether migration is worth it.
Clinical AI / airway systems / emergency medicine