A platform to upload, explore, and connect knowledge across PDFs with advanced search, contextual insights, and podcast-style content consumption.
- Bulk Upload: Upload multiple PDFs at once (represents past documents you’ve read).
- Fresh Upload: Open an additional new PDF (represents the current document you’re reading).
- High-Fidelity Display: Render PDFs with accurate formatting (PDF Embed API preferred).
- Relevant Sections: Identify & highlight up to 5 highly relevant sections across PDFs.
- Section Definition: Headings + their content (logical chunks of the document).
- Snippet Preview: Show 2–4 sentence extracts from relevant sections (like search-engine results).
- Deep Linking: Clicking a snippet opens the corresponding PDF and jumps to the exact section.
- Fast Retrieval: Related sections/snippets load quickly for seamless user engagement.
- Efficient Ingestion: Past documents are processed within earlier round performance limits.
- LLM-powered insights across PDFs.
- Summaries, contextual connections, and knowledge synthesis from multiple documents.
- Smart Search – Search across all your documents with contextual snippets.
- Podcast – Generate and listen to audio summaries for your PDFs.
- Insights – AI-powered insights that connect the dots across your reading history.
- Frontend: React (TypeScript + TSX)
- Backend: GoLang & Python
- Database: PostgreSQL
- Deployment: Docker (command below)
docker run -e ADOBE_EMBED_API_KEY=<> -e GOOGLE_API_KEY=<> -p 8080:8080 -p 8081:8081 <image_name> ADOBE embeded api : 5da540d215fd4f1a9cd89f97a45bae83In case of some any issue with docker please consider this video
https://drive.google.com/drive/folders/1BDh66AfRQ__DbGhgpxGt4lYS0AZfLHY-?usp=sharing