Healthcare Data Platform
A synthetic, PHI-free healthcare payer platform: Data Vault 2.0 integration underneath, Kimball dimensional consumption on top, surfaced here as read-only marts. Every number on this site traces to a gold or platinum model — the app layer carries zero business logic, it only dials the marts to the glass.
Patients (clinical)
Resolved Persons
Members
Claims
Total Paid
Two ways into the data
🩺 Clinical
The provider's view of the population — encounters, conditions, medications, and demographics. Start with the clinical overview below.
💳 Payer
The plan's view — identity, enrollment, and claims.
- Person 360 — one patient, both sides, resolved on one spine
- Member Months — the enrollment heartbeat + coverage mix
- Claims — volume, dollars, and in-network status
Same data, a role-designed screen
The marts above are analyst views — the whole population, read as a report. The persona screen dials the same resolved-person data into a screen someone uses in a role, one member at a time:
- Member Services call screen — the "rep on the phone" view:
pick the caller, then work down identity → coverage → money → recent activity. Same
person_360+person_eventsmarts as Person 360, reframed as a product a customer-service rep would actually use on a call. - Clinical exam-room screen — the provider's view: pick the
patient, then read down identity → clinical picture → recent care → care over time.
The same
person_360+person_eventsmarts, reframed clinically — conditions, medications, and encounter history, the way a nurse or provider reviews a chart before a visit.
Behind the build
- Feature Board — the Review Cockpit — the interactive board used to plan and review this project, embedded live in the dashboard (a frozen snapshot as of the last deploy). Kanban, roadmap, status, bugs, decisions, and the git-derived cockpit — click through it right here.
Clinical overview
Encounters
Conditions
Prescriptions
Clinical Claim Cost