Campus AI workflow
PilotChecks a dissertation against the graduate formatting rules, fixes what it safely can, and flags the rest for the student to decide.
- 84Rules checked13 categories
- 17+Auto-fixesCommon issues
- 3Result bucketsFixed, action, verify
Accountability
How this use case is governed
- Service owner
- TritonAI solutions team and the sponsoring graduate-education service owner
- Human oversight
- The student and graduate-education reviewer retain responsibility for the final document.
- Measurement plan
- Rules checked, issues resolved, review cycles, time to acceptance, and false findings.
- Data boundary
- Public description; submitted documents follow approved service controls
Product media
See the workflow in action
Silent screen recording showing document upload, automated rule checks, auto-fixes, and the fixed/action/verify results breakdown.
- GEPA Rules
- Auto-Corrections
- Review Report
How it works
How the workflow fits
The problem it addresses, what it actually does, and how far along it is.
Problem
There are a lot of dissertation formatting rules, and checking them by hand pulls students and reviewers away from the actual work.
Solution
The pilot checks a Word document against 84 UCSD GEPA formatting rules across 13 categories and applies 17+ automatic fixes for the common problems. Findings land in one of three buckets: fixed, action needed, or verify. Each one comes with an explanation, so the student can see what changed and why.
Current status
Pilot. It does not replace the official submission requirements or the final graduate-education review. Students upload a .docx, pick the document type, run the checks, and download the result.