ProveNN

The description states that ProveNN:

• Verifies that an invoice submitted for expense reimbursement hasn't been altered since it was issued.
• Hashes each invoice with SHA-256 and stamps it with a unique reference code and QR code upon creation.
• When an employee submits the invoice, the platform extracts the reference code, recomputes the hash, and reports match, mismatch, or not found.
• Allows company admins to approve or reject submissions independently of the automated result.
• Exports approved invoices to Excel for reimbursement processing.
• Provides a cross-tenant view for platform operators without exposing one company's data to another.

Inference The system is built around cryptographic verification as a core function, not as part of a broader expense management suite. It focuses on detecting document tampering rather than preventing it or managing the full reimbursement workflow.

The description states:

• The platform aims to close a gap in current expense fraud practices where companies still rely on manual human review.
• Expense reimbursement fraud is described as a measurable cost with median losses of $50,000 per incident.
• It is positioned as an instant and automatic solution for verifying document integrity.

Inference The positioning is focused on solving a specific problem — document tampering in expense reimbursement — rather than offering a full suite of expense management tools. The authors emphasize that the system does not aim to replace human decision-making but to automate verification.

The description states:

• ProveNN is intended for companies that process expense reimbursements.
• It targets employees submitting invoices and company admins reviewing submissions.
• A separate platform console allows operators (platform admins) to view usage across partners or companies.

Inference The target customer appears to be mid-to-large enterprises with formalized expense reimbursement processes. The system is designed for integration into existing workflows, not as a standalone tool.

Not evidenced.

The description does not include any information about pricing, licensing, monetization strategy, or business model.

The description states:

• Built with Go (chi router), Next.js, TypeScript.
• Uses JWT for platform users and API keys for SDK integrations.
• PostgreSQL stores structured data; MinIO holds original PDFs.
• Background worker stamps reference code and QR using pdfcpu and computes hash asynchronously.
• Verification decodes QR with gozxing, falling back to text search.
• Exports generated with excelize.
• Instrumented with Prometheus and Grafana.

Inference The architecture is modular and built with a clear separation of concerns. It uses standard tools for backend (Go), frontend (Next.js), storage (PostgreSQL, MinIO), and monitoring (Prometheus, Grafana). The system is designed to be scalable and observable from the start.

Not evidenced.

There is no evidence of revenue, customers, or adoption beyond the hackathon. The authors state that the core cryptographic verification is complete and tested end-to-end, but there is no mention of real-world deployment or usage.

Not evidenced.

The description does not provide any information about competitors or market positioning relative to existing solutions for expense reimbursement fraud detection.

• No evidence of traction: The system was built as a hackathon project and is not reported to be in production.
• Unverified claims: The description makes strong claims about fraud prevention but does not provide data or validation.
• Limited scope: The current version only supports digital-native PDFs, not scanned receipts or OCR-based verification.
• No pricing or monetization model: There is no indication of how the product will be monetized.

1. Has ProveNN been tested in any real-world environment beyond the hackathon?
2. What are the specific use cases or industries where this system has been deployed?
3. Are there any known limitations or edge cases with the current verification process (e.g., how it handles legitimate reprocessing)?
4. How does the platform handle integration with existing expense management systems?
5. Is there a plan to support OCR for scanned receipts, and what is the timeline for that feature?
6. What are the plans for monetization and customer acquisition?