Medallion architecture

RebelCore organises data work into four stages. You do not need to understand the underlying data platform to use them; the stages simply identify where data is in the preparation process.

Import source data  →  Create a module  →  Curate and build  →  Ask Agent

The terms Bronze, Silver, and Gold may appear in the product and documentation:

Stage	RebelCore action	Outcome
Bronze	Import files	Source files are available in a dataset batch.
Silver	Create a module	The imported data is organised and labelled.
Gold	Curate the Tree and build a semantic dataset	A focused data selection is ready for Agent.
Agent	Start a session and ask questions	Agent answers within the selected access, exposure, and privacy controls.

1. Import source data

Open Datasets, click the + button, name the batch, add Excel or CSV files, and upload them.

At this point, the source files have been received but are not ready for a project.

Import files

2. Create a module

When the uploaded dataset is ready, select CREATE MODULE. Review the data, add the required labels, select an LLM exposure level, and build the module.

The completed module can now be selected when creating a project.

Create your dataset

3. Curate and build

Create or open a project, then use the Tree to:

• Navigate the dataset structure.
• Select the fields relevant to the intended questions.
• Review available data suggestions.
• Build one or more semantic datasets.

How the Tree works

4. Ask Agent

Open RebelCore Agent, select the semantic datasets required for the task, choose a response mode, set the privacy controls, and submit your question.

RebelCore Agent

Access at each stage

Roles can separate these responsibilities:

• A data steward can import and build modules.
• A project analyst can curate projects and semantic datasets.
• An Agent user can ask questions without receiving data-administration access.
• The Customer Super Admin can review activity and manage access.

See Governance & access and Roles & permissions.

Fix problems at the right stage

• A file or column is missing: review the Datasets import.
• The labels or module are wrong: rebuild the module.
• The Tree includes irrelevant fields: adjust the Tree selection.
• Agent cannot find the expected information: confirm that the correct semantic dataset is ready and selected.
• Agent returns less detail than expected: check the effective LLM exposure level.