SchemaSentry

SchemaSentry is a web app (with optional CLI agent) that continuously monitors data schemas across warehouses and key SaaS sources and alerts teams when changes will break downstream jobs. It connects to Snowflake/BigQuery/Databricks plus common ingestion tools, fingerprints tables/views, and compares changes against dbt models and BI semantic layers. When drift is detected (column rename/type change/nullability shifts), it generates impact analysis: which dbt models, dashboards, and APIs are likely to fail, and suggests safe fixes (e.g., add cast, backfill, update model). This is a combination traditional + AI app: deterministic diffing for accuracy, with AI only for readable explanations and proposed patches. The product is intentionally narrow: schema drift is a constant, expensive source of incidents, and existing observability suites are often overkill and priced for enterprises.

← Back to idea list