DataADO moves files across SFTP, cloud and on-prem systems, runs SQL-based ETL across SQL Server, Oracle, PostgreSQL and MySQL, maps what's inside unfamiliar databases, and orchestrates all of it as reliable, auditable workflows. All of it self-hosted, so nothing leaves your network.
Move files securely across SFTP, FTP/FTPS, S3, Azure Blob & Data Lake, SMB and local systems — with checksum verification, retries and streamed PGP encryption built in.
Export query results to CSV or Parquet, load CSV, TSV, XLSX, JSON or Parquet straight into a table, and run scripts — across SQL Server, Oracle, PostgreSQL and MySQL, in either direction. Streamed, so file size is never the limit.
Compose activities into a directed graph with parallel branches, loops, conditions, per-node retries and idempotent replay — started by a schedule, a file arrival or a webhook.
Point DataADO at an unfamiliar database and it maps what's inside — classifying columns by their data, not their names — so you know exactly where to extract from.
Move many tables between instances — same engine or across SQL Server, Oracle, PostgreSQL and MySQL. Pick the tables, and see what will break before anything moves: folded identifiers, types that cannot survive the trip, and the ones DataADO refuses outright. Keys, indexes and same-engine procedures come across with the data.
Build top-to-bottom flows with per-activity forms, connection pickers, loop & condition boxes, and a live JSON view.
Run on a cron schedule, the moment a file lands, or from an inbound token-authed webhook.
Read Parquet straight into a database table, and export any query back out to Parquet with real column types, not stringified CSV. CSV, TSV, XLSX and JSON import too. Memory stays bounded either way, so the row count is not the limit.
Encrypt and decrypt multi-gigabyte files with bounded, predictable memory — proven past 2 GB.
Discover PAMI, documents, file-pointer columns, flags and lookup tables on any of the four engines — classified by value shape, grouped into business concepts, and turned into a draft extraction query.
Describe a pipeline and get a validated draft workflow; explain a failed run in plain language. Inference runs on a model server you host — prompts never leave your network.
Load between SQL Server, Oracle, PostgreSQL and MySQL in either direction, with type mapping and schema pre-flight handled for you.
Every change is hash-chained and append-only — enforced at the database itself, not just the app.
Fail-closed isolation, five server-side roles, and Microsoft Entra sign-in — including per-tenant B2B directories.
Health checks, stuck-run detection and reaping, and failure alerts to email or Teams.
Every activity records what its ${…} expressions resolved to, how many rows or files it moved, and the database's own error code on failure — ORA-… or the SQL error number. Secrets redacted; no full-payload noise.
▶ Click through the interactive demo — every screen, on sample data, nothing to install →
Handed an unfamiliar clinical or ERP system? DataADO scans it and tells you where the demographics, orders, results, documents, flags and code tables actually live — even when the tables are named T_47 and the columns C1…C7. The same map comes back whichever of the four engines the source runs on.
DataADO's AI is an opt-in add-on that runs entirely on a model server you host. Prompts carry run logs, schemas and connection codes — so they never touch a third-party API.
Say what you need moved, in plain language. DataADO drafts a real workflow against your own connection codes, checks it with the same validators that guard a hand-built one, and opens it in the visual designer for you to change.
Tell us what you're moving or the database you need to understand, and we'll show you the flow.
Contact us →