Documentation, cited.
A multi-site provider deployed a clinical documentation assistant in its own cloud. Answers cite their sources, and retrieval follows each clinician's access rights.
Answers clinicians could not verify.
Clinicians spent time searching across systems for prior notes, lab results, and care plans. An initial AI assistant gave fluent answers, but they could not be checked against the record and did not reliably respect patient access permissions.
For a healthcare system, an ungrounded answer is not a minor defect. It is a clinical risk and a compliance exposure at once.
Retrieval that inherits access, not just relevance.
We built permissions into retrieval. Each query carries the clinician's access scope, and answers cite the note or result they use. Answers without a supporting citation are suppressed.
This meant building the evaluation harness around groundedness from day one: for every test question, we measured not just whether the answer was correct, but whether it was traceable to a real, permitted source.
Deployed inside the provider's own cloud.
The retrieval index, model calls, and citation store run inside the provider's cloud. Clinical data stays within its governed environment, and access permissions are checked on every query.
Trusted enough to use daily.
Clinicians adopted the assistant for daily documentation lookup once the citations made answers checkable in seconds rather than minutes. Compliance signed off once the access-control re-checks were demonstrated against real audit scenarios, not just described in a document.