Jobs
An onboarding job runs an onboarding configuration against a datasource.
Jobs are asynchronous. Create a job, then retrieve it to monitor status, progress, current stage, and the dataset produced by the run.
Create an initial job
An initial onboarding job creates a new dataset.
The job references the datasource, onboarding configuration, and processing compute configuration. It also names the dataset that will be created.
Create an incremental job
An incremental onboarding job adds new data to an existing dataset.
Incremental jobs use a newer datasource and can include an increment_label to identify the update.
Monitor jobs
Retrieve jobs to monitor onboarding progress or inspect previous runs.
To inspect a specific job, retrieve it by id.
Job responses include the job status, overall progress, current stage, and stage progress.
Initial job stages
An initial onboarding job can move through these stages:
- Data validation
- Masking
- Analysis
- Statistics
- Preprocessing
- Sharding
- Transformation
- Initialisation
- Persistence
- Teardown
Incremental job stages
An incremental onboarding job can move through these stages:
- Masking
- Analysis
- Statistics
- Preprocessing
- Sharding
- Transformation
- Metadata
- Persistence
- Teardown
The masking stage is only used when unseen category handling is enabled.
Update a job
Update an onboarding job when you need to rename it.
Delete a job
Delete an onboarding job when it should no longer be kept.

