Description
Ask questions about any PDF directly inside Telegram—powered by AI and vector search. This n8n workflow processes uploaded documents, stores them in a vector database, and intelligently answers your queries based on the content.
🧠 How It Works:
✅ 📥 Telegram Bot Upload – Users send a PDF document to your Telegram bot ✅ 📄 PDF Parsing – The file is parsed and chunked for semantic understanding ✅ 🧠 Vector Embedding & Storage – Document chunks are embedded and stored in Pinecone (or your preferred vector DB) ✅ 💬 Query Detection – If a user sends a message without a document, the bot treats it as a question ✅ 🤖 AI-Powered Answering – The bot searches the vector store and returns an accurate answer using the context
🔍 It Automates:
✅ Accepting PDFs through Telegram uploads ✅ Extracting and chunking text from the PDF file ✅ Creating embeddings using a selected LLM model ✅ Storing document vectors in Pinecone ✅ Searching stored PDFs when a user asks a follow-up question ✅ Generating context-aware answers from relevant chunks
💡 Why Choose This Workflow:
✅ Query any PDF in seconds with natural language ✅ No need for a frontend—use Telegram as the interface ✅ Lightweight, fast, and fully customizable ✅ Easily switch out vector stores or LLM models ✅ Ideal for documentation, research, contracts, manuals, and more
👤 Who Is This For:
✅ Teams dealing with complex documents or manuals ✅ Researchers and students needing instant PDF lookup ✅ Professionals managing multiple contracts or legal files ✅ Educators sharing reading material with interactive querying ✅ Anyone wanting an AI assistant that understands PDFs
🔗 Integrations:
✅ Telegram Bot API (via n8n Telegram Trigger node) ✅ Pinecone Vector Store (for document embeddings) ✅ Groq LLM API (can be swapped with OpenAI, Claude, etc.) ✅ PDF Parsing Tool (integrated via n8n’s Function/HTTP nodes)
🔧 Setup Instructions:
📲 Create a bot via @BotFather on Telegram and get the API token 📩 Add the Telegram Trigger node in n8n to receive messages 📄 Configure PDF parsing logic with file buffer 🧠 Embed content into Pinecone (or Supabase/Postgres/Qdrant) 🤖 Route user text messages through your chosen LLM 📤 Return answers directly to Telegram in real-time
📚 Your Personal AI PDF Expert Inside Telegram From contracts to course notes, this Telegram chatbot brings PDFs to life with instant, smart Q&A—no app-switching, just upload, ask, and get answers.
Link : https://lovable.dev/projects/7d071541-4aca-4c8f-9311-c194b8e0451b
