Lo scenario
A healthcare AI team needs to model how SDLC artifacts (model cards, data cards, evaluation reports) are tracked, linked to compliance requirements, audited, and included in regulatory submissions.
Cosa contiene questo disegno
Leggi le decisioni che ne stanno alla base.
01
Introduced a SubmissionArtifact join table to support many-to-many between RegulatorySubmission and Artifact.
02
Separated AuditEvent and ComplianceCheck to distinguish continuous audit logging from one-time compliance verification.
03
Linked Artifact to Project, ArtifactType, and User to capture ownership, categorization, and creation context.
Use this schema when building a regulated AI SDLC system that must maintain artifact provenance, demonstrate compliance, and produce regulatory submission records.