Datasets

Datasets are the fundamental building blocks of Catalyzed. They represent a logical collection of data tables that can be queried, transformed, and analyzed.

What is a Dataset?

A dataset in Catalyzed is a container that:

• Belongs to a specific team
• Contains one or more tables (data sources)
• Has defined schemas per table
• Can be queried using SQL via the Query Engine

Dataset Tables

Within a dataset, data is organized into tables. Each table:

• Has a schema defining its columns and types
• Supports schema evolution and migrations
• Maintains statistics for query optimization
• Can be indexed for faster lookups

Creating a Dataset

Datasets can be created through the UI or API:

curl -X POST https://api.catalyzed.ai/datasets \
  -H "Authorization: Bearer YOUR_API_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "teamId": "YOUR_TEAM_ID",
    "name": "my-dataset",
    "description": "My first dataset"
  }'

Querying Datasets

Once data is loaded into a dataset, you can query it using SQL through the Query Engine. See the Query Engine documentation for details.

API Reference

See the Datasets API for complete endpoint documentation.