Taleshape OÜ releases Shaper, an open-source tool enabling developers to build interactive data dashboards using SQL queries and DuckDB without complex infrastructure.
For developers and data analysts tired of wrestling with heavyweight BI tools, a new open-source solution has emerged that simplifies data visualization through a radically straightforward approach. Shaper, released by Estonia-based Taleshape OÜ, enables users to create shareable dashboards using nothing but SQL queries – powered entirely by DuckDB's embedded analytical engine.
At its core, Shaper addresses the friction in traditional data workflows where analysts often need to export query results to separate visualization tools. By combining DuckDB's in-process OLAP capabilities with a reactive UI framework, Shaper renders visualizations directly from SQL statements. Users write queries like SELECT date, revenue FROM sales, and Shaper automatically generates interactive charts, tables, and filters without manual mapping.
Screenshot: Shaper's interface showing SQL editor alongside generated visualizations
The architecture leverages DuckDB's in-memory processing to handle transformations locally, eliminating external database dependencies for smaller datasets. For larger datasets, Shaper connects to PostgreSQL, MySQL, and other databases via its connector system. Dashboards can be shared as standalone links or embedded in applications, with live refresh capabilities when underlying data changes.
Deployment options highlight Shaper's developer-centric approach:
- Local prototyping: Run instantly via Docker (
docker run -p5454:5454 taleshape/shaper) with zero configuration - Production hosting: Kubernetes-ready with configurable authentication and caching
- Managed cloud: Taleshape offers hosted instances with monitoring and priority support
What sets Shaper apart is its avoidance of proprietary visualization languages. Unlike tools requiring custom syntax (e.g., Vega-Lite or Metabase's MBQL), Shaper uses raw SQL as its interface. This lowers adoption barriers for teams already fluent in SQL while providing escape hatches for complex scenarios through DuckDB's extensions for geospatial, JSON, and full-text search.
The project adopts the business-source-friendly Mozilla Public License 2.0, with core functionality remaining perpetually open-source. Commercial offerings include enterprise features like SAML authentication and dedicated SLAs.
For teams seeking lightweight alternatives to monolithic BI platforms, Shaper represents an intriguing middle ground between writing custom dashboards from scratch and adopting feature-bloated solutions. Its success will depend on how effectively it scales beyond ad-hoc analysis – an area the team plans to address through upcoming collaboration features.
Getting started:
Comments
Please log in or register to join the discussion