Uxie

Built with Next.js, TypeScript, Tailwind CSS, shadcn/ui components, tRPC for type-safe APIs, Prisma ORM for database management, Supabase for backend services, Vercel AI SDK and Langchain for AI features, Pinecone for vector embeddings, Next Auth for authentication, and Liveblocks for real-time collaboration, this full-stack application transforms static PDFs into interactive learning experiences. The platform features comprehensive PDF annotation tools with highlighting, underlining, and margin notes, intelligent note-taking with custom editor blocks and AI-powered text autocompletion, AI chat functionality that answers questions about document content using RAG (Retrieval Augmented Generation), automatic flashcard generation from document content with LLM-generated feedback on answers, sentence-by-sentence text-to-speech with visual highlighting (English only), OCR support for extracting text from scanned documents and images, real-time collaborative editing powered by Liveblocks allowing multiple users to annotate simultaneously, and vector embedding storage in Pinecone for semantic search and AI context. The AI integration uses the Vercel AI SDK and Langchain to provide contextual understanding of PDFs, enabling students to ask questions and receive answers grounded in the actual document content.

This application is invaluable for students who want to enhance their study sessions with AI-powered learning tools, researchers annotating academic papers and collaborating on document analysis, professionals reviewing contracts or reports with team members, educators creating interactive reading materials with embedded questions and flashcards, and anyone who learns better through active engagement rather than passive reading. Uxie demonstrates how to build AI-enhanced document viewers, implement real-time collaboration in React applications, integrate vector databases for semantic search, create custom text editors with AI features, and design engaging learning experiences through technology.