An AI-assisted compliance auditing platform designed to help federal auditors detect improper vendor billing through document ingestion, extraction, reconciliation, exception flagging, decision capture, and report tracking.
These screens show the core audit product experience: upload, dashboard review, improper payment analysis, archives, settings, and full analysis workflows.
The journey artifact shows how user actions and system/AI responsibilities move across upload, extraction, comparison, decision, and tracking.
Federal auditors need AI support that speeds up extraction and reconciliation without hiding the decision logic. The product concept separates the AI pipeline from human decision ownership so auditors can understand what was flagged, why it matters, and how to act on it.
The challenge was to present AI-generated findings, vendor billing data, contract comparison, and exception status in a way that supports trust, review, and defensible decisions.
Connected document upload, AI extraction, reconciliation, exception flagging, reporting, archives, and settings into one product workflow.
Designed AI as a support layer that surfaces discrepancies while preserving human review, decision capture, and accountability.
Structured vendor bills, pricing schedules, flags, decisions, and archived reports so teams can review and defend audit outcomes.
Leadership signal: This work demonstrates senior product judgment in an AI-enabled compliance environment: workflow design, data interpretation, risk framing, human-in-the-loop interaction, and production-minded UX for complex federal operations.
The system helps extract and compare data, but the interface makes the human decision path visible.
Each stage is designed as a clear step in a workflow rather than a black-box AI process.
The screens prioritize traceability, status, and next action so compliance work stays auditable.