Vibe Trading and Ollama: stock dashboard setup

This tutorial builds a Python stock analysis dashboard around Vibe Trading and Ollama. The presenter describes Vibe Trading as an open source workflow that turns financial questions into executable code and backtests. The overview covers Indian market support, configurable fees, research history and scheduled reports. Security and data leakage protections are claims made by the presenter, rather than independently verified results.

The dashboard has two tabs: Stock Copilot and AI Research Engine. Stock Copilot uses yfinance to fetch price history and metrics, including the P/E ratio and 52-week range. It plots 50-day and 200-day moving averages, then sends those statistics to a local LLM through Ollama. The walkthrough names Gemma 2B for report generation. The app displays the Markdown report, saves it to output.md and provides a download button.

The research tab passes a natural-language query to the Vibe Trading command-line tool and streams its output into the dashboard. Setup includes Python dependencies, a downloaded Ollama model and a running Ollama server. The presenter launches the app with streamlit run app.py and demonstrates Reliance.NS. Google Antigravity is the editor used in the walkthrough.

Although the source title describes offline operation, the dashboard fetches market data through yfinance. Local report generation does not establish that the complete workflow works without a network connection.