Welcome to AskMyDocs — an AI-powered document chatbot that allows you to upload documents and interact with them conversationally. Built using Retrieval-Augmented Generation (RAG), it combines document retrieval, embeddings, and large language models to deliver contextual answers directly from your files.

👉 Click the image above to watch the full demo video.

AskMyDocs simplifies document analysis using AI:

• Upload multiple files
• Ask questions naturally
• Receive contextual AI-generated answers
• Quickly extract insights without manual searching

This project demonstrates how modern AI pipelines can transform document interaction and knowledge retrieval.

• Supports PDF, TXT, CSV, DOCX files
• Automatic text extraction
• Efficient document chunking and embedding

• Conversational Q&A from documents
• Context-aware responses
• Retrieval-Augmented Generation pipeline

• Optional prompt refinement
• Document summarization
• Fast inference powered by Groq LLM

• Streamlit interactive dashboard
• Simple upload & chat workflow
• Clean, responsive design

.
├── icons/
│   └── ico.png
├── src/
│   ├── analytics.py
│   ├── app.py
│   ├── conversation.py
│   ├── document_utils.py
│   ├── prompt_refiner.py
│   └── text_processing.py
├── README.md
├── pyproject.toml
├── setup.sh
└── uv.lock

• Python 3.9+
• UV package manager recommended

Install uv:

pip install uv

Create .env file in root:

GROQ_API_KEY=your_api_key_here

Required for AI inference.

git clone https://github.com/hindav/AskMyDocs.git
cd AskMyDocs

Using uv:

uv run streamlit run src/app.py

Or:

bash setup.sh

Open browser:

http://localhost:8501

1. Upload documents
2. System converts them into embeddings
3. Ask questions naturally
4. Get contextual AI responses

Optional:

• Enable prompt refinement
• Generate document summaries

• Python
• Streamlit
• LangChain
• FAISS Vector Database
• HuggingFace Embeddings
• Groq LLM (Llama models)

• Research document analysis
• Study material Q&A
• Knowledge base assistant
• Business document exploration
• Personal document AI assistant

Hindav Deshmukh AI • Data Engineering • Machine Learning

GitHub: https://github.com/hindav LinkedIn: https://www.linkedin.com/in/hindav/

MIT License — free to use, modify, and distribute with attribution.

⭐ If you find this project useful, consider starring the repository!