The Integrations hub
Where connectors, pipelines, and automations all live now.
Integrations is the single place where your data plumbing lives. Connectors, pipelines, and automations all share one page now — no more hunting across separate sidebar sections to find them.
If an article tells you to "open Integrations," this is the page it means.
One page, four tabs
Open Integrations from the left sidebar and you'll land on a tabbed view:
| Tab | What it shows |
|---|---|
| All | Everything together — pipelines, automations, and connectors. |
| Pipelines | Just your pipelines. |
| Automations | Just your automations. |
| Connectors | Your saved connections, plus Add connection to create a new one. |
The page remembers the tab you were last on, so if you mostly live in Pipelines, that's where you'll land next time.
"Create integration" starts everything
There's one button — Create integration, top right — and it opens a short wizard instead of scattering a different "create" button on every tab. It asks "What do you want to do?" and offers three cards:
| Card | What you get |
|---|---|
| Sync data | Copy data into your lakehouse on a schedule. This is the pipeline flow. |
| Automate a task | Chain steps — pipelines, SQL, notebooks, semantic models — into one workflow. |
| Set up a live connection (Beta) | Query the source directly from SQL, with nothing to sync first. |
Sync data walks you through source → destination → name and then drops you into the pipeline configuration. See Build your first pipeline.
Automate a task goes to the automation builder. See What automations do.
Set up a live connection takes you to your lakehouse with the add-live-source panel open. It's disabled unless the project's lakehouse can host it — Trino or Doris. See Live connections.
Creating a connection starts from the Connectors tab's own Add connection button, which opens the connector gallery. See Build your first connector.
Connectors had their own sidebar section once. They moved here — and the old sidebar spot now nudges you toward Integrations if you go looking. Pipelines and automations consolidated the same way.
Why it's organized this way
Pipelines, automations, and connectors are three parts of one job: get data in, move it, and orchestrate the work around it. Keeping them on one surface means you can see the whole picture — what's connected, what's running, and what's failing — without bouncing between pages. The All tab is the fastest way to take that in at a glance.