One graph from your database to your dbt models.
GetMyData pulls from your databases and APIs, lands it where you want it, and runs your own dbt project against it — scheduled, monitored and audited. One tool, one permission model, and your extracted data never touches our servers.
9 sources · 6 destinations · 8 dbt warehouses · single sign-on included
Three things that usually need three tools.
Extraction, transformation and orchestration behind one login, one role model and one audit log — instead of stitched together across vendors.
Extract from a database or an API and land it in storage or your warehouse. Write SQL, reuse a saved query, or pick a table and let GetMyData build the query for you.
Point at your git repository and get scheduled jobs, run history, lineage, test results with failing rows, and docs — running your own dbt Core, not a fork of it.
Run an extraction, then a dbt build, on a schedule. Say what happens on success and on failure, and get told the moment something breaks.
Every source, destination and warehouse is available today, with no per-connector charge — you are never billed for turning one on.
Your data stays yours.
The questions your security team will ask, answered by how the product is built rather than by policy.
Extracted data is written to your storage and your warehouse. GetMyData holds the schedule, the run log and the metadata — not the rows.
Every workspace gets its own database with its own credentials, not a shared table with a tenant column.
Runners connect outbound only. When your warehouse sits inside your network, run a self-hosted runner and nothing needs to be exposed.
Sign up and your first account is a local one. Once you are in, connect Entra ID, Google Workspace or any OIDC provider and your team signs in through your own directory.
AES-256-GCM secret store, or bring your own Azure Key Vault. Secrets are masked in every log, error and query preview.
A queryable, exportable audit log, retained for as long as your plan allows, plus per-run logs you can stream live while a job is running.
Managed runners
We host the compute. Nothing to install, nothing to patch, and you can scale up to more replicas when a backlog builds, then let them sleep when it does not.
Metered in runner minutes — job execution time, not uptime, so an idle runner costs nothing. Prepaid, never overage: running out pauses new managed work instead of billing you more.
Self-hosted runners
Run the agent inside your own network, next to the databases it reads. It dials out to us — there is no inbound firewall rule and no VPN to maintain.
Self-hosted work consumes no runner minutes at all, so a workload you host yourself costs nothing beyond your plan.
See it against your own data.
Connect a source and run your first extraction in a few minutes — or have us walk you through it.