I’m building a progressive web application that lets teams capture, store and retrieve documents from any device, even when they’re offline. The core workflow is simple: a user takes a photo or uploads a file, the system runs printed-text OCR on it, and the resulting PDF plus extracted text are filed so that others can search or tag them later. Multi-user logic matters. We only need two roles—Admin and Editor. Admins can create or deactivate users, adjust global settings and purge or export the full archive. Editors handle day-to-day scanning, tagging and updating of documents; they’ll never see system settings or user management screens. Because people will rely on this while moving between office and field, it has to feel like a native mobile app: installable from the browser, offline caching, and instant push or local notifications that remind Editors to scan new paperwork at scheduled times they define themselves. When they reconnect, the queue should sync automatically. Key technical expectations • Modern PWA framework (React, Vue or Angular) with Service Workers and IndexedDB for offline data. • OCR engine optimised for printed text—Tesseract.js or a comparable web-friendly library is fine as long as accuracy is solid and processing happens client-side or via a lightweight serverless function. • REST or GraphQL API, JWT auth and role-based access control. • Responsive UI that meets Lighthouse PWA audits. Deliverables 1. Source code in a public or private Git repo. 2. Build scripts and deployment guide for a standard Docker or serverless setup. 3. A brief README explaining how to add new Admin or Editor accounts, set reminder schedules and adjust OCR language packs. 4. Final build running on a demo URL so I can test on phone and desktop before sign-off. I’ll provide Figma screens and a sample set of documents once we start; you plug in the tech, wire up the OCR and make sure reminders fire reliably.