Requests & Self-Service
As the Connected AI Platform Team, our goal is to provide a scalable, transparent, and efficient way to process use case–specific platform requests.
There are two complementary mechanisms for changing or extending your Managed Kubeflow environment:
- Service Requests — structured, form-based requests (via ServiceNow / My BMW IT) for common, well-defined actions: provisioning a workspace, managing users, setting up CI Trust or a CDH connection.
- PR-based Self-Service — a Pull Request to our Infrastructure-as-Code products repository, for changes that are not (yet) covered by a Service Request.
Which one should I use?
| Service Requests | PR-based Self-Service | |
|---|---|---|
| How | Submit a form via ServiceNow / My BMW IT | Open a Pull Request to the products repository |
| When to use | For anything covered by an existing Service Request (workspace provisioning, user management, CI Trust, CDH connection, …) | Only for changes not (yet) available as a Service Request or in the Self-Service Portal |
| Who executes the change | The Connected AI Platform team, triggered automatically by your request | You, by editing your product's product.yaml (or related file) and opening a PR; the Connected AI Platform team reviews and merges |
| Turnaround | Automated / queued for the platform team | Depends on PR review by the platform team |
Always prefer Service Requests
All actions that can be triggered via a Service Request must be executed via the Service Request process. The PR-based Self-Service should only be used for actions that are currently not available through the Self-Service Portal or Service Request workflow.
Start with Service Requests — only fall back to PR-based Self-Service if your request isn't covered there.
What's possible
Via Service Requests
- 🆕 New Managed Kubeflow Workspace — provision a new Managed Kubeflow Workspace.
- 👥 Manage CAIP Space Users — add or remove users from your CAIP Space.
- 🔐 Establish CI Trust — connect a GitHub repository to your Managed Kubeflow Workspace.
- 🔗 Establish CDH Connection — connect your Managed Kubeflow Workspace to a CDH Environment.
- ❓ General Inquiry / Other Request — anything else not listed above.
Directly via CDH
- 👥 Add/Remove CDH Datasets — grant or revoke dataset access on the CDH side once a CDH Environment is linked to your namespace.
Via PR-based Self-Service
- 🔐 Add/Remove Custom IAM Policies — attach custom IAM policies to CAIP default roles.
- 🔁 Add/Remove Repository in CI-Trust — add or remove a repository from the CI-Trust authentication flow.
- 🧪 Enable MLflow Experiments & Model Registry — turn on MLflow for your namespace.
- ⚡ Enable EMR-on-EKS — run Spark jobs on your namespace.
- 📊 Enable Grafana — turn on a Grafana instance for your namespace.
- 🔒 Integrate with CDH PII API — encrypt/decrypt PII data (VIN ↔ VAN).
- 🚗 Integrate Vehicle Shadow — enable the Vehicle Shadow data stream.
- 🗄️ Enable Vector Database — turn on AWS PGVector for your namespace.
- 🚀 Enable KServe — turn on model serving with KServe.
Anything not listed above may still be possible — check with the Connected AI Platform team via a general inquiry.