Generate realistic datasets for demos, learning, and dashboards. Instantly preview data, export as CSV or SQL, and explore with Metabase.

Features:

• Conversational prompt builder: choose business type, schema, row count, and more
• Real-time data preview in the browser
• Export as CSV (single file or multi-table ZIP) or as SQL inserts
• One-click Metabase launch for data exploration

• Next.js (App Router, TypeScript)
• Tailwind CSS + ShadCN UI (modern, dark-themed UI)
• OpenAI API (GPT-4o for data generation)
• Metabase (Dockerized, launched on demand)

1. Clone the repo:

   git clone <your-repo-url>
   cd dataset-generator

2. Create your .env file:

   Copy the example file and fill in your OpenAI API key:

   cp .env.example .env.local

   Then edit .env.local and add your OpenAI API key after the = sign.

3. Start the Next.js app:

4. Generate a dataset:

   • Use the prompt builder to define your dataset.
   • Click "Preview Data" to see a sample.

5. Export or Explore:

   • Download your dataset as CSV or SQL Inserts.
   • Click "Start Metabase" to spin up Metabase in Docker.
   • Once Metabase is ready, click "Open Metabase" to explore your data.
   • When done, click "Stop Metabase" to shut down and clean up Docker containers.

• /app/page.tsx – Main UI and prompt builder
• /app/api/generate/route.ts – Synthetic data generator (OpenAI)
• /app/api/metabase/start|stop|status/route.ts – Docker orchestration for Metabase
• /lib/export/ – CSV/SQL export logic
• /docker-compose.yml – Used only for Metabase, not for the app itself

When you click "Start Metabase", it will launch Metabase in a Docker container. Once ready:

1. Click "Open Metabase" to access the Metabase interface
2. Follow Metabase's setup process
3. To analyze your generated data:

Key Points:

• You only pay for the preview/spec generation (~$0.05 per preview)
• All downloads use the same columns/spec, just with more rows, and are free

• When you preview a dataset, the app uses OpenAI to generate a detailed data spec (schema, business rules, event logic) for your chosen business type and parameters.
• All actual data rows are generated locally using Faker, based on the LLM-generated spec.
• Downloading or exporting data never calls OpenAI again—it's instant and free.

1. Select your business type, schema, and other parameters.
2. Click "Preview Data" to generate a 10-row sample (incurs a small OpenAI cost).
3. Download CSV/SQL for as many rows as you want—no extra cost, always uses the same schema/columns as the preview.

• One Big Table (OBT): A single, denormalized table with all relevant columns.
• Star Schema: Multiple tables (fact + dimension) for more advanced analytics. The LLM spec guides the structure, and the generator outputs all tables locally.

• To add new business types, rules, or schema logic, edit lib/spec-prompts.ts