An interface that pushes new records to consumers the instant they arrive, rather than making them poll for updates.
A streaming API keeps a connection open and delivers each record the moment it is produced, so a consumer reacts to events as they happen instead of repeatedly asking whether anything is new. Common transports include WebSockets and server-sent events, and it is the natural delivery model for real-time data reaching a trading desk or a live model.
The design questions are about guarantees, not plumbing. Does the stream deliver each record at least once, exactly once, or best-effort; does it preserve ordering; and what happens when a slow consumer falls behind - does the server buffer, drop, or apply backpressure. A good streaming interface answers these explicitly and pairs the live stream with a bulk or REST path for history, so the same data serves both a live desk and a backtest.
A live evaluation measures EdgeOrigin against your coverage requirements: decision-ready real-time data, delivery latency into your systems, and record-level provenance.