Managed application artifacts
Pipeline applications are currently created and provisioned with Redgold during the managed preview. The checked-in artifact set forms the durable customer boundary.
Artifact set
| Artifact | Responsibility |
|---|---|
| Rust serde schema | Defines request, response, and persisted record shapes |
pipelines/*.toml | Declares routes, execution stages, datasets, and capabilities |
| Rust transform crate | Contains domain-specific behavior executed by the managed runtime |
| Inline tests and fixtures | Demonstrate the expected transformation and contract behavior |
| Application contract | Summarizes routes, types, storage, permissions, and current limits |
These files stay together in Git so a behavior change can be reviewed as one unit. An AI coding agent can propose the schema, manifest, transform, and test; the customer controls the resulting diff and approval.
Division of responsibility
The customer reviews application meaning and behavior. Redgold validates the artifact graph, provisions datasets, checks route and capability policy, pins transform artifacts, and operates deployment and runtime infrastructure.
During the preview, a typical change moves through this sequence:
- Agree on the request, stored record, and expected response.
- Review the proposed schema, pipeline TOML, transform code, and inline test.
- Approve one Git change containing the complete contract.
- Redgold validates, provisions, and deploys the approved revision.
- Verify one bounded request against the managed route.
The pipeline application walkthrough shows a complete illustrative manifest and transform. Validate a pipeline application change describes the review checklist, and verify a managed deployment covers the final smoke request.
Future expansion
Self-service scaffolding and publication, portable releases, application catalogs, declarative client surfaces, sharing, and offline installation build on the same checked-in contract. Their current component boundaries and planned work are documented in the Pipeline Apps overview.
Build a vertical
How to build an application vertical on the platform, in order — Rust schema, ingestion, transforms, service surface, UI, then deployment.
Import your data
Getting an existing dataset — CSV, JSONL, Parquet, or a Postgres table — into the platform, what works today, and how to verify the write landed.