Metabase developer Matthew Hefferon created an open source AI Dataset Generator to solve the problem of finding realistic test data for demos. The tool combines LLM-generated schemas with local data generation using Faker.js, offering configurable business types, growth patterns, and export options. It gained significant traction with 600+ GitHub stars and contributions from the community, providing a cost-effective alternative to manually searching datasets or relying solely on ChatGPT for data generation.

4m read timeFrom metabase.com
Post cover image
Table of contents
Why not Kaggle or ChatGPTWhat I actually didHow It Works

Sort: