A data pipeline is the set of steps that move data from its sources to a destination, transforming it along the way. Pipelines run in batch or in streaming mode and are usually coordinated by an orchestration tool that schedules and retries each step.
A retailer might run a nightly pipeline that pulls sales records, cleans and aggregates them, and loads the result into a warehouse for reporting.
Delivering data on schedule is not the same as making it ready for AI. A pipeline can move records reliably yet leave out the lineage and state needed to reproduce a model result later. AI-ready transformation keeps that traceability intact, so the output can be replayed, not just delivered.