Company
Databricks
20 cases from Databricks (General, SaaS B2B, Fintech), each digested from a public talk or engineering post and linked to the original. 4 report results in production.
Tasks: Answer questions over data, Make metric definitions consistent, Evaluate models, Explain a metric change, Analyze user behavior and 7 more. Methods: Semantic layer, LLM-as-a-judge, Golden set / benchmark, Text-to-SQL, MCP, Vector / semantic search and 48 more. Tools: Databricks Genie, Unity Catalog, Unity Catalog Metric Views, Databricks SQL, Databricks Genie Code, MLflow and 48 more.
Tech blog: databricks.com