φ(ai) PHI AI
Menu
Clinical workspace
Patients & charts PHI AI assistant Signature queue Ambient documentation Imaging worklist Summarization Inbox triage Patient instructions
Revenue & records
Claims & billing Prior auth & appeals Documentation gaps Release of information Sensitive categories
Population
Cohort builder No-show risk Scheduling optimization Trial pre-screening Quality measures Ingest & mapping QA Data Store
Integration
Source & target EMRs Source patient import Bulk import manager Streaming data Bulk export manager Streaming export
System
Control panel Model monitoring
Governance
Registry & preflight Fairness screen Ambient consent gate Source attributes EMR conformance Retention Audit
Documentation
System documentation Implementation & setup guide Feature documentation System architecture EMR connections Emulators & non-PHI setup Compliance & responsibility Attribution
Ryan GomezRole
Signed in — Ryan Gomez · System Administrator
You stay on this screen — each role sees it as their own permissions allow. The assistant thread and the patient in context start fresh.
Working archetypes — combinations people really hold
One role at a time — each role on its own
DocumentationDocs
System documentation Implementation & setup guide Feature documentation System architecture EMR connections Emulators & non-PHI setup Compliance & responsibility Attribution
Repository

Data Store

A live audit of everything the Data Store holds: the inventory by category, its time coverage, and integrity checks re-computed on this page load. Every category the platform can house is represented — clinical, financial, and behavioral health.

Inventory

33,311
patients (named, synthetic)
150,009
encounters
1922-08-21 → 2026-08-29
78,093
conditions
1919-02-09 → 2026-08-30
47,571
medication orders (deduplicated)
1923-09-16 → 2026-08-30
6,884
allergy records
1918-06-21 → 2026-12-28
55,748
immunizations
1945-03-06 → 2026-12-28
26,367
procedures
1931-05-07 → 2026-12-28
440,889
labs & vitals
1922-08-14 → 2026-08-29
38,413
clinical notes
1931-01-05 → 2026-08-30
59,347
adjudicated claims (financial)
1929-04-30 → 2026-08-30
123,171
claim adjudication lines
1,756
imaging studies
1925-12-11 → 2026-08-29
30
psychotherapy notes (gated)
2025-11-01 → 2026-01-25
5
release-of-information requests
2026-07-30 → 2026-08-20
0
audit events (hash-chained)

Integrity checks

CheckResultVerdict
Referential integrity
Encounters, conditions and claims whose patient does not exist.
 0 orphaned rows pass
Patients without encounters
A patient with no visit history is a capture-completeness question.
 0 of 33,311 pass
Claims without adjudication lines
A claim total with no line detail cannot be reviewed line by line.
 0 of 59,347 pass
MRN uniqueness
Duplicate identifiers are how records merge wrongly.
 0 duplicates pass
Audit chain
Every event re-hashed against its predecessor.
 0 events verified pass
Sensitivity classification sweep
Unflagged notes whose text suggests a sensitive category — the platform treats an unclassifiable record as excluded and counted.
 0 candidates for review pass

Sensitive-category distribution

15,029
Mental health — withheld from ordinary retrieval
2,385
SUD — 42 CFR Part 2 — withheld from ordinary retrieval
1,914
Reproductive health — withheld from ordinary retrieval
142
HIV — withheld from ordinary retrieval
30
Psychotherapy notes — separate store, dedicated role
What this mirrors. In the deployable platform the system of record is an encrypted object store; the queryable index, the analytics layer and every screen here are derived from it and rebuildable, and a scheduled reconciliation proves the derived structures against the store. This page is that discipline made visible: the inventory and every check above are recomputed from the Data Store on each load, never cached, never asserted.