data-context-extractor

Generating custom data analysis skills involves extracting domain knowledge from data analysts to document schemas, disambiguate entities, and define metrics. This captures company-specific terminology and query patterns for data warehouses like BigQuery, Snowflake, or Databricks.

Documenting tribal knowledge for a BigQuery data warehouse requires extracting domain knowledge directly from analysts. This involves schema discovery, entity disambiguation, and defining metrics to generate reference files that guide AI data exploration.

Yes, you can add new metrics and terminology to an existing data analysis skill using an iteration mode. This allows you to document new domains and query patterns, adding context to the skill without starting the schema discovery process over.

Yes, this data context extraction approach works with Databricks. It supports schema discovery, metric definition, and query pattern documentation across various data warehouses, enabling you to build tailored analysis skills for your specific Databricks environment.

The best way to define company-specific metrics like ARR for AI data analysis is through a guided extraction process. By interviewing analysts and discovering schemas, you can establish clear metric definitions and document query patterns in reference files.