I have an AI-focused Python concept that I now want packaged into a clean Streamlit web app. The core idea is to run one or more autonomous “agents” that process data and return insights in real time; from there, the user should be able to trigger classification, regression, or clustering logic (whichever best fits the uploaded dataset) and immediately view results inside the same interface. Here’s what I need from you: • A lightweight, well-structured Python codebase that spins up a Streamlit app, loads the agent logic, and exposes controls for the three model modes mentioned above. • Clear separation between data ingest, model inference, and UI layers so I can swap in new models later. • Simple, attractive visualisations for predictions and any metrics the agent returns. • README with setup steps, required packages, and guidance on how to extend or retrain the underlying models. The deliverable is a fully functional repo I can run with st run, plus the documentation. If anything here is unclear, let me know and I’ll clarify before you begin coding.