Integrate with KFP server: list runs and pipeline information directly from Notebook
- Dominant language
- Python
- Stars
- 703
- Forks
- 161
- Avg merge
- 6d 3h
- Merged PRs (30d)
- 10
Description
## Summary
Kale currently submits pipelines to KFP but provides no way to browse existing pipelines or runs from within JupyterLab. Users must switch to the KFP UI for any post-submission visibility.
Eder will create the epic.
**Blocked by:** [kubeflow/sdk#125](https://github.com/kubeflow/sdk/issues/125) — Kubeflow SDK Pipelines integration must land first.
## Proposal
Surface KFP server information directly in JupyterLab — list pipelines, browse runs, view status, inspect parameters and artifacts — without leaving the notebook.
## Open questions
- Should this use the KFP API directly or wait for the Kubeflow SDK Pipelines client?
- What is the minimal useful view — run list with status, or full artifact/parameter inspection?
Contributor guide
Research direction
No files, tests, or entry points are named. Start by checking the blocked kubeflow/sdk#125 dependency and the existing Kale pipeline-submission integration, then resolve whether to use the KFP API or SDK and define the minimum view; done means JupyterLab can browse pipelines and runs with status, parameters, and artifacts as scoped.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- jupyter-notebook, python
- Domain
- frontend, machine-learning
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Active
- Clarity
- Needs clarification
- Newbie friendliness
- 30/100