K2-k2/DocuGenie

🧞‍♂️ DocuGenie: AI-Powered Document Intelligence

DocuGenie is a professional-grade Retrieval-Augmented Generation (RAG) platform that allows users to have interactive, multi-lingual conversations with their documents. Built using Streamlit, LangChain, and Google Gemini 2.5 Flash-Lite, it transforms static files into dynamic knowledge bases.

🌟 Overview

DocuGenie bridges the gap between raw data and actionable insights. By combining vector embeddings with state-of-the-art language models, the app can answer specific questions, summarize complex reports, and find exact page references within seconds.

🛠️ Key Features

• —Multi-Format Support: Seamlessly process PDF, DOCX, PPTX, XLSX, TXT, and Markdown files.
• —Multilingual Support: Communicate with your data in over 15 languages, including English, Hindi, Telugu, Spanish, and more.
• —Smart Retrieval (RAG): Uses ChromaDB and Google Generative AI Embeddings to find the most relevant context before answering.
• —Contextual Accuracy: Provides specific source citations (page numbers) for its answers to ensure transparency.
• —Usage Management: Built-in daily limits for uploads and messages to manage API costs and prevent abuse.
• —Persistent UX: Features chat history management, downloadable transcripts, and suggested actions (Summarize, Key Takeaways, etc.).

🚀 Deployment on Hugging Face Spaces

1. 1.SDK: Select Streamlit.
2. 2.Secret Keys: Navigate to Settings > Variables and Secrets*.
3. 3.Add GOOGLE_API_KEY_N2 with your Google Gemini API Key.
4. 4.Requirements: Your requirements.txt should include:

    streamlit
    langchain-google-genai
    langchain-community
    langchain-chroma
    unstructured[all-docs]
    chromadb
    pypdf

🏗️ Technical Architecture

• —Orchestration: LangChain (Chains & Document Loaders)
• —Vector Database: ChromaDB (Local persistent storage)
• —LLM: Google Gemini 2.5 Flash-Lite (Optimized for speed and efficiency)
• —Embeddings: Google gemini-embedding-001
• —UI/UX: Custom CSS-injected Streamlit components

💻 Local Setup

# 1. Clone the repository
git clone <your-repo-url>

# 2. Install dependencies
pip install -r requirements.txt

# 3. Set Environment Variable
export GOOGLE_API_KEY_N2="your_api_key_here"

# 4. Run the app
streamlit run app.py

Developed with ❤️ by Kesar Agarwal

Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference