A smart PDF question-answering application that lets you upload PDF documents and ask questions about their content using AI. Built with React, Express, and Google Gemini AI.
Explore the application and chat with your own PDF:
- PDF Upload & Processing - Upload PDF files and extract text content
- AI-Powered Q&A - Ask questions about your PDFs and get intelligent answers
- Session Management - Create up to 3 separate chat sessions for different documents
- Chat History - Keep track of your conversations with persistent storage
- Vector Search - Find relevant content from your documents using embeddings
- Real-time Interface - Modern, responsive chat interface with loading states
- React 19 with TypeScript
- Vite for fast development and building
- TailwindCSS for styling
- Axios for API calls
- React Router for navigation
- React Toastify for notifications
- Lucide React for icons
- Express.js with TypeScript
- LangChain for document processing and AI workflows
- Google Gemini AI for text generation and embeddings
- Neon PostgreSQL for vector storage
- Redis for session and chat history management
- Multer for file uploads
- PDF-Parse for PDF text extraction
-
Upload PDF: The user uploads a PDF document through the interface.
-
Text Extraction: The server processes the PDF and extracts its textual content.
-
Text Chunking: The extracted text is split into smaller, manageable chunks to optimize processing.
-
Vector Embeddings: Each chunk is converted into vector embeddings using Google AI Embedding APIs.
-
Storage:
- PostgreSQL stores the vector embeddings.
- Redis manages the real-time chat history and session data.
-
Question Processing: User-submitted questions are passed through a LangChain-powered workflow.
-
Context Retrieval: Relevant chunks are retrieved via vector similarity search against the stored embeddings.
-
Answer Generation: Google Gemini generates a contextual response using the retrieved chunks and chat history.
