README.md

metadata

title: Smart Doc QA
emoji: 📄
colorFrom: blue
colorTo: indigo
sdk: docker
app_port: 8501
pinned: false
short_description: RAG-based document Q&A with LangChain, ChromaDB & Groq

smart-doc-qa

A RAG-based document Q&A system. Upload any PDF and ask questions in natural language — get accurate, source-grounded answers.

Features

• Upload any PDF document
• Ask questions in natural language
• Get answers grounded in document content (no hallucination)
• Powered by semantic search + LLM

Tech Stack

• LangChain — RAG orchestration
• ChromaDB — vector database
• HuggingFace — embeddings (sentence-transformers)
• Groq — LLM inference (Llama 3.1)
• FastAPI — REST API (coming soon)

Status

Currently in active development.

Setup

```bash git clone https://github.com/T0N-M0Y/smart-doc-qa cd docu-query python -m venv venv source venv/bin/activate pip install -r requirements.txt ```

👤 Author

Md. Reja E Rabbi Tonmoy — ML Engineer LinkedIn • GitHub