Hi HN, we built Riftur to automate the repetitive work involved in document gap analysis. In many roles and industries, people still compare requirements to drafts manually—switching between two documents, marking gaps, and tracking mismatches in spreadsheets. We wanted to see whether we could handle that comparison step with AI in a way that’s transparent and useful.

Riftur takes a requirement set (proposal instructions, audit criteria, evaluation rubrics, regulatory checklists, internal standards, etc.), interprets each item, and compares it to a document you upload. It highlights what’s missing, partially addressed, or inconsistent, so users don’t have to manually read two documents side-by-side to figure out alignment.

The tool isn’t limited to compliance—you can also use it for things like mapping internal engineering guidelines, reviewing technical documentation, checking process adherence, or assessing structured submissions in education or training contexts.

The current version is live and testable, and we’re trying to understand how well the gap-identification approach generalizes across different types of documents.

We’d appreciate feedback on: – whether the gap identification feels accurate and explainable – the UX around understanding mismatches – formats or frameworks you’d want us to support – any concerns or limitations you think we should be aware of

Happy to answer technical questions about the approach and implementation.