Python Desktop Document Chatbot

Job ID: 40043208

Budget: $30 – $250 USD

I need a self-contained desktop app, built entirely in Python, that lets me drop files into a designated folder and then chat with the content offline—essentially a local version of ChatGPT limited to my own documents.

The reader must ingest and index PDF, Word (docx), Text files, and Excel sheets as soon as they appear in the folder. From there, a simple chat window should let me ask questions in natural language and receive accurate, context-aware answers pulled only from those local sources. No external API calls, no cloud storage—everything has to run on my machine.

I am comfortable launching from an executable or using a single-command script, but setup should be minimal. Feel free to use PyQt, Tkinter, or another lightweight GUI framework and whichever NLP stack you prefer (LangChain, llama-cpp-python, sentence-transformers, etc.) as long as the final package is truly offline and open for me to review or extend later.

Acceptance criteria:
• Drag-or-drop (or auto-watch) folder for PDF, docx, txt, xlsx
• Instant indexing with re-index on file change
• Chat interface with streaming or near-instant answers sourced only from the indexed data
• Clear README plus all source code in one repo or folder
• Runs on Windows 10 without additional paid dependencies

Hand over the compiled app and full Python code when done.