Skip to main content

Create a dashboard with AI

This is the core of Vibedasher: you describe what you want, and an AI agent does the heavy lifting — generating SQL queries and visualization code on the fly to produce a live, interactive dashboard.

Start a dashboard​

  1. From the app, start a new visualization and pick the dataset you prepared in Connect a data source.
  2. In the chat panel, describe what you want in plain language. For example:

    "Show monthly revenue and order count for the last 12 months, with a KPI for total revenue and a bar chart broken down by product category."

  3. The agent generates the SQL and the chart configuration, then renders the result. Watch the dashboard build up live as it works.

Refine by chatting​

You don't have to get the prompt perfect the first time. Keep the conversation going:

  • "Switch the bar chart to a line chart."
  • "Add a date range filter that applies to every chart."
  • "Break revenue down by region instead of category."
  • "Add a period-over-period comparison to the revenue KPI."

The agent edits the existing dashboard in place rather than starting over.

What the agent can build​

  • Chart types — bar, line, area, scatter, pie, and KPI cards, rendered with Apache ECharts.
  • Multi-page dashboards — organize related charts across pages.
  • Filters & selectors — date ranges, dimension filters, and measure switchers.
  • Cross-chart interactions — cross-filtering and highlighting between charts.
  • Advanced measures — period-over-period comparisons, comparison periods, and dynamic measure switching.
tip

Be specific about the measure (what to count or sum), the dimension (how to break it down), and the time range. The clearer the ask, the better the first result.

Next step​

Happy with the dashboard? Publish and share it.