An AI-powered application that allows users to upload PDF documents and ask questions about their content.
The app uses Retrieval-Augmented Generation (RAG) to search document text and generate answers using a language model.
🔗 Live App: https://rag-pdf-chatbot.streamlit.app
🔗 GitHub Repository: https://github.com/Littajosethottam/rag-pdf-chatbot
- Upload one or multiple PDF documents
- Ask questions about document content
- Semantic search using embeddings
- Vector similarity search with FAISS
- AI-generated answers based on document context
- Source citation showing the document and page number
- Interactive chat interface built with Streamlit
The application follows a Retrieval-Augmented Generation (RAG) pipeline:
Instead of sending the entire document to the language model, the system retrieves only the most relevant sections and uses them as context to produce accurate answers.
Many organizations and researchers work with large collections of documents such as reports, research papers, manuals, and policy documents. Searching through these files manually can be inefficient and time-consuming.
This project explores how AI-powered document assistants can transform static PDFs into interactive knowledge systems. By combining vector search with language models, users can quickly access insights from complex documents through simple natural language queries.
The project demonstrates how modern AI systems can support knowledge discovery, document understanding, and decision support.
- Summarizing research papers
- Extracting key concepts from study materials
- Supporting exam preparation
- Searching internal reports and documentation
- Extracting insights from strategy or financial documents
- Supporting data-driven decision making
- Turning document collections into searchable AI assistants
- Supporting internal documentation systems
- Quickly locating information in regulatory or policy documents
- Extracting relevant clauses from large legal texts
- Python
- Streamlit
- OpenAI API
- FAISS (Vector Similarity Search)
- PyPDF
- NumPy
Clone the repository: git clone https://github.com/Littajosethottam/rag-pdf-chatbot.git
cd rag-pdf-chatbot
Install dependencies:
pip install -r requirements.txt
Run the application:
streamlit run app.py
Create a .streamlit/secrets.toml file and add your OpenAI API key:
OPENAI_API_KEY = "your-api-key"
You can ask questions such as:
- Summarize this document
- What are the key concepts discussed?
- Explain the methodology used
- What conclusions does the paper draw?
- List the main findings of this report
This project showcases several key AI engineering concepts:
- Retrieval-Augmented Generation (RAG)
- Vector embeddings for semantic search
- Vector databases using FAISS
- AI-powered document analysis
- Interactive AI applications using Streamlit
Litta Jose Thottam
LinkedIn
https://linkedin.com/in/litta-thottam-12a223193
GitHub
https://github.com/Littajosethottam
This project is intended for educational and demonstration purposes.