Data Product Builder
What is the Data Product Builder?
The Data Product Builder is the control plane for managing the lifecycle of a data product — from repository initialization to governed version releases. It connects your descriptor (DPDS) to Git, enforces validation, and ensures every published version is traceable, immutable, and compliant.
If you haven’t created a data product yet, start from the Data Product Catalog to browse or create one, then open its Builder to manage it.
Key concepts
- Data product — A governed data asset defined by a descriptor compliant with the Data Product Descriptor Specification (DPDS) . It includes identity, ownership, domain, interface components (ports), lifecycle stages, and governance metadata.
- Descriptor (DPDS file) — The document stored in Git that defines the data product. It is the single source of truth for the product’s identity, contracts, metadata, and policies.
- Repository — The Git repository linked to the data product. It stores the descriptor and may also contain code, pipelines, policies, or infrastructure definitions.
- Version (release) — An immutable snapshot of the descriptor, identified by a Git tag (e.g. 1.2.0). Each version guarantees traceability and compatibility for consumers.
Init, Build & Publish
Data product management follows three main phases. All of them are available from the Builder page:
Connect the data product to a Git repository and publish it for the first time. You can:
- Create a new repository directly from Blindata
- Connect an existing repository from your Git provider
Each data product must have a descriptor file (DPDS) stored in the repository. This descriptor becomes the authoritative definition of the product. Best practice: initialize and define the descriptor before exposing the product to consumers.
2. Descriptor editing (shift-left governance)
Edit the descriptor directly in Blindata and commit changes to Git. This is where the data product owner defines:
- Identity and domain
- Ownership and responsibilities
- Interface components (ports) and contracts
- Governance metadata
- Terms of use and semantics
This phase implements shift-left governance: contracts and metadata are defined before marketplace publication. Early specification ensures discoverability, compliance, and accountability.
3. Version creation (release management)
When ready, create a new version: tag the repository (e.g. 1.1.0), add a changelog, and publish the release. The Git tag freezes the exact state of the descriptor at that point in time, ensuring:
- Full traceability
- Consumer compatibility
- Auditable lifecycle evolution
The Builder displays all registered versions so you can browse, search, and create new ones.
The Builder page
The Builder is the operational hub for a single data product. Access it from the Data Product Catalog by selecting a data product and switching to the Builder view.

What you’ll see on the Builder
| Section | What it does |
|---|---|
| DevOps Stages | Shows the lifecycle across environments (e.g. dev, test, prod) and the deployed version in each. Links to the DevOps module for lifecycle management. |
| Release Versions | Lists all published versions, highlights the latest release, and shows validation status. Create new releases from here. |
| Git Repository | Displays the connected repository: branch, tags, latest commit, and author. Use EDIT to open the descriptor editor. |
| General Info | Preview of the descriptor (fully qualified name, domain, version, description) from the selected branch. |
To configure or connect a Git repository, see Platform Administration .
Descriptor Editor
Press EDIT to open the descriptor editor.

The editor lets you:
- Modify the data product descriptor file
- Validate structure and compliance before committing
- Commit changes directly to the linked repository
Validation ensures the descriptor complies with the Data Product Descriptor Specification before changes are committed. The editor bridges governance and DevOps: business metadata and technical definitions live together in version-controlled infrastructure.
Data Product Descriptor Specification (DPDS)
Blindata data products follow the Data Product Descriptor Specification (DPDS) . DPDS defines the standard structure of a descriptor, including:
- Fully qualified name (FQN)
- Ownership and domain
- Versioning
- Interface components (ports)
- Lifecycle stages
- Governance metadata
Descriptors are validated automatically to ensure consistency and compliance. Standardization through DPDS enables interoperability across teams and platforms.
Blindata Observer for ODM
If you use Open Data Mesh (ODM) as your data platform, the Blindata Observer synchronizes data product updates, metadata, and lifecycle information into Blindata. It listens for ODM platform events and performs actions such as:
Data product events
- DATA_PRODUCT_UPLOAD — Register a data product and assign ownership
- DATA_PRODUCT_REMOVAL — Remove the data product from Blindata
Version and metadata events
- DATA_PRODUCT_VERSION_UPLOAD — Upload metadata for ports and assets
- POLICIES_UPLOAD — Upload policy evaluation results
Lifecycle events
- STAGES_UPLOAD — Synchronize lifecycle stages across environments
For setup and configuration, see the Blindata Observer for ODM GitHub Repository .
Permissions
To work with the Builder, you need the following platform roles:
| Action | Required roles |
|---|---|
| View and interact | Agent Viewer and Data Ops Viewer — required to see the Builder section and browse its content. |
| Modify, create, and save | Agent Editor and Data Ops Editor — required to edit the descriptor, create releases, and save changes. |
In addition, responsibilities and team policy over the single data product or domain must be configured according to your tenant settings. Access to modify a given data product or domain depends on these assignments.