SchemaDrift

SchemaDrift is a web app (with a lightweight agent) that monitors NoSQL collections/tables for schema drift and risky changes over time. It continuously samples documents/rows, infers field shapes, types, optionality, and nested structures, then versions these “observed schemas” per environment. When a deploy or data pipeline introduces breaking drift (e.g., type flips, missing required fields, cardinality explosions, oversized documents), it alerts with concrete examples and a diff you can hand to the team. It also generates validation rules and migration suggestions for popular stacks (e.g., MongoDB JSON Schema, DynamoDB item constraints via app-layer validators). This is not a silver bullet: inference can miss rare edge cases, and you’ll need to tune sampling and privacy controls. But it gives teams a practical early-warning system for the messy reality of schemaless databases.