Summarize, Translate, Merge, Split, Compress & Convert — all powered by AI, processed securely in your browser.
DocFlow AI is a modern, full-stack web application that provides AI-powered document processing tools — similar to iLovePDF but enhanced with artificial intelligence. All file processing happens in-memory for maximum privacy and speed.
| Role | Name |
|---|---|
| AI & Backend | m.shri subashini |
| Frontend & UI | k.prakash |
- AI Summarize — Upload text/documents → get AI-generated summaries
- AI Translate — Translate document content into any target language
- AI Question Generator — Generate study/quiz questions from documents
- OCR Text Extraction — Extract text from scanned images using Tesseract.js
- Merge PDF — Combine multiple PDF files into one
- Split PDF — Extract specific page ranges into separate files
- Compress PDF — Reduce PDF file size while maintaining quality
- Add Watermark — Stamp custom text watermarks on all PDF pages (adjustable opacity)
- Image ↔ PDF — Convert images to PDF and vice versa
- Compress Image — Reduce image file size with quality control
- Email + Password sign up with email verification
- Google OAuth sign-in (one-click)
- Protected routes — all tools require authentication
- Dark/Light mode toggle
- Responsive design — mobile-friendly with collapsible sidebar
- Drag & drop file upload with validation
- Processing animations and toast notifications
| Feature | Status |
|---|---|
| PDF → Word Conversion | 🔶 Beta |
| Word → PDF Conversion | 🔶 Beta |
| Excel → PDF Conversion | 🔶 Beta |
| PPT → PDF Conversion | 🔶 Beta |
| Remove Watermark | 🔶 Beta |
| Edit PDF (inline text) | 🔶 Beta |
These features are marked with a BETA badge in the UI.
| Layer | Technology |
|---|---|
| Frontend | React 18, TypeScript, Vite 5 |
| Styling | Tailwind CSS 3, shadcn/ui, Framer Motion |
| Backend | Supabase Edge Functions |
| AI | Google Gemini / OpenAI via AI Gateway |
| Auth | Supabase Auth + Google OAuth |
| PDF Processing | pdf-lib (client-side) |
| OCR | Tesseract.js (client-side) |
| Image Processing | browser-image-compression |
| File Handling | JSZip, FileSaver.js |
User uploads file
→ Client-side validation (type + size ≤ 20MB)
→ Process in-memory (pdf-lib / Tesseract / AI API)
→ Return result for download
→ Files are NEVER permanently stored
src/
├── components/ # Reusable UI components
│ ├── Sidebar.tsx # Navigation with tool categories
│ ├── HomeView.tsx # Dashboard with tool cards
│ └── FileDropZone.tsx
├── contexts/ # Auth context provider
├── pages/ # Route pages (Index, Auth, ResetPassword)
├── tools/ # Individual tool implementations
│ ├── AISummarize.tsx
│ ├── AITranslate.tsx
│ ├── AIQuestions.tsx
│ ├── OCRTool.tsx
│ ├── MergePDF.tsx
│ ├── SplitPDF.tsx
│ ├── CompressPDF.tsx
│ ├── AddWatermark.tsx
│ ├── CompressImage.tsx
│ └── ConvertTool.tsx
├── lib/ # Utilities and API helpers
└── integrations/ # Backend client configuration
supabase/
└── functions/ # Edge Functions (AI endpoints)
- No permanent file storage — all processing in browser memory
- Email verification required for account creation
- Google OAuth for secure one-click sign-in
- Protected routes — authentication required for all tools
- Sign Up / Sign In — Email (verified) or Google OAuth
- Dashboard — Browse tools by category (AI, Organize, Editing, Convert)
- Select a Tool — Click any tool card
- Upload File — Drag & drop or click (max 20MB)
- Process — Click the action button
- Download — Get your result instantly
git clone <https://github.com/shri8977/hackathon-hustle>
cd docflow-ai
npm install
npm run devRuns on http://localhost:5173
MIT License
Built with ❤️ using React + TypeScript + AI