What problem does it solve?
Collaborate with the AI statistician to produce a polished data dashboard from a topic brief, research question, or set of DST table references. This workflow drives the end-to-end process from data exploration to dashboard validation and visual inspection, guided by a startup prompt describing the subject and desired dashboard concept.
Core Features & Use Cases
- Iterative dashboard design workflow: data exploration → dashboard creation → validation → visual inspection → iteration.
- Startable via a startup prompt and a persistent interactive CLI session (uv run python -m varro.playground.cli --user-id 1).
- Step-by-step guidance for exploring data, selecting relevant tables, determining dimensions and time ranges.
- Ability to commission dashboards with a clear name, outputs, layout, filters, and narrative framing.
- Validation and iteration loop to fix SQL, charts, labels, defaults, and narrative text.
- Final validation and recording of results, including trajectory and dashboard URL.
Quick Start
Provide a startup prompt describing the subject, data angles, and dashboard concept, then launch the interactive CLI session.