jaivardhan2409/document-ai-analyst
0
๐ง Document AI Analyst โ Enterprise Agentic RAG System
Upload complex PDFs, DOCX files, CSVs, or text documents and chat with an AI agent that provides accurate, cited insights powered by Retrieval-Augmented Generation.
โจ Features
- Multi-Format Upload โ PDF, DOCX, TXT, CSV, and Markdown with smart chunking
- Semantic Search โ High-precision vector retrieval using Google Gemini embeddings
- Streaming Chat โ Real-time AI responses with inline source citations
- Data Isolation โ Per-user knowledge bases and vector collections for complete privacy
- Persistent Sessions โ Seamless authentication with secure JWT tokens
- Premium UI โ Custom dark-themed Streamlit interface with glassmorphism and micro-animations
๐๏ธ Architecture
๐ Quick Start
- Clone the repository and create your
.envfile:
cp .env.example .env- Add your Gemini API Key to the
.envfile. - Start the application using Docker Compose:
docker compose up -d --build- Access the UI at
http://localhost:7860. Register an account, upload a document, and start chatting!
๐ง Local Development
Backend
cd backend
python3 -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt
uvicorn app.main:app --port 8000 --reloadFrontend
cd frontend
python3 -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt
streamlit run app.py๐ฆ Environment Variables
๐ ๏ธ Tech Stack
Built with: FastAPI โข Streamlit โข ChromaDB โข Google Gemini โข PostgreSQL โข Docker Compose
