Goal
Enable maintainers (and users) to track all jobs run on a given score set, view the results and status of each job, understand why annotations may be missing from certain variants, and display overall pipeline progress in the UI.
Key Requirements
- Each job run is recorded with metadata.
- Jobs can be grouped by some mechanism to support a progress view in a user interface.
- Per-variant annotation status and missing reason are tracked in a dedicated table.
- Efficient understanding of the 'latest' job run, and the 'current' annotation statuses + reasons for variants
- API endpoints allow querying jobs by correlation_id, viewing job details, and listing per-variant annotation status and reasons.
- Support for polling/streaming of status into the UI for API driven updates of pipeline progress.
Benefits
- Maintainers can audit all jobs and their outcomes for any score set.
- Users can see real-time or near-real-time progress of their pipelines.
- Missing annotations are explained at the variant level, improving transparency and troubleshooting.
Goal
Enable maintainers (and users) to track all jobs run on a given score set, view the results and status of each job, understand why annotations may be missing from certain variants, and display overall pipeline progress in the UI.
Key Requirements
Benefits