How nareshedagotti/document-query-system works
Project type and purpose
• This is a Python‐based, single‐process full-stack application that provides a web interface (via Gradio) for uploading documents (PDF/DOCX/TXT), extracting their text, and querying them using two external LLM services (Groq API and Ollama).
File structure
• Top-level files:
• app.py ‑- application entry point, launches the Gradio server and ties everything together
• ui.py ‑- defines Gradio components (upload widget, radio button for model selection, chat history, text input)
• utils.py ‑- contains text-extraction routines (PDF, DOCX, TXT)
• models.py ‑- wraps the API calls to Groq and Ollama, implements send/query/receive logic
• requirements.txt, README.md, LICENSE, GitHub workflows for CI/CD
README for additional insights
• Documents the user flow: Upload → Select model → Ask question → Instant chat response.
• Calls out external dependencies: Groq API key, Ollama endpoint.
• No persistent database – state (chat history) is kept in memory.