Sapion Health — an AI operating layer for clinical research

01The problem

Clinical research runs on operational fragmentation. Study startup lives in documents, pre-screening in spreadsheets, consent on paper or a portal, regulatory files in another system entirely — and enrollment, the step every trial timeline depends on, sits at the mercy of all of them. The people who could benefit most from trials often never hear about them; Sapion's founding conviction is that access to breakthroughs shouldn't depend on a zip code.

02What we're building

With Sapion, the work is an AI operating layer that treats trial operations as one connected flow rather than a stack of point tools:

• Study startup — the documentation-heavy front end of every trial, structured instead of re-typed.
• Patient pre-screening and enrollment, including a live pre-screening tool already in use — matching intake against study criteria before staff time is spent.
• Consent workflows that stay auditable end to end.
• Regulatory documentation connected to the study state that produces it, not reconstructed after the fact.
• Trial management views tying the thread together through a shared AI layer.

03How we're working

Research operations are regulated territory, so the build leads with the guardrails: human accountability at clinical decisions, audit trails by default, and AI applied where it compresses paperwork — never where it replaces judgment. It's the tiered-autonomy pattern from our practice, applied to a domain where the tiers really matter.

04Where it stands

In private preview, with early-access components — including the pre-screening tool — already live with users, and the platform expanding module by module alongside Sapion's team.