Skip to content
DataViz Storyteller

Get started

1. Clone and install

Section titled “1. Clone and install”
Terminal window
git clone https://github.com/SpandanM110/DataViz.git
cd DataViz
npm install

2. Configure Tambo

Section titled “2. Configure Tambo”

Run npx tambo init or manually create .env.local from example.env.local and set:

NEXT_PUBLIC_TAMBO_API_KEY=your-api-key

3. Run the app

Section titled “3. Run the app”
Terminal window
npm run dev

Open http://localhost:3000 to use the app.

Optional: Local LLM

Section titled “Optional: Local LLM”

Set NEXT_PUBLIC_TAMBO_URL (or TAMBO_LOCAL_URL) in .env.local to your Tambo API URL (e.g. for Qwen). The app proxy tries that first, then falls back to Tambo Cloud.

Optional: Notebook code execution

Section titled “Optional: Notebook code execution”

Code blocks in the notebook panel have a Run button. The server runs Python in a project venv (run-code-venv); missing packages (matplotlib, pandas, numpy, scikit-learn, etc.) are auto-installed on first use. Upload a CSV and ask to summarize or plot — the dataset is sent with your message so the model can analyze it.

Next steps

Section titled “Next steps”