Skip to main content

​
What We Store

  • Connection metadata — Connection names, types, and organisation association. Not credentials (those are encrypted).
  • Pipeline configuration — Source, destination, stream selection, sync mode, write mode, transformation rules.
  • Sync state — Cursor values and CDC positions for incremental and CDC syncs.
  • Run metadata — Run IDs, status, row counts, timestamps, error messages. Not the actual synced data.
  • User and organisation data — Account info, team membership, roles.

​
What We Do Not Store

  • Your source data — Data flows from your source through the ETL service to your destination. We do not persist it.
  • Credentials in logs — Credentials are never written to logs. Only connection IDs and metadata may appear.
  • Synced data — The platform does not retain a copy of the data you sync. It exists only in your source and destination.

​
Data Flow

Source → ETL (transient) → Destination. The ETL service reads from the source, applies transformations, and writes to the destination. Data is held in memory only during the sync.