Agentic Experiences25% of the exam
Draft with Einstein, Suggest Relationships and Data Pro
How Tableau Next's generative AI authoring tools build semantic model components: Draft with Einstein for calculated fields, AI Semantic Description Generator (Beta) for descriptions, Suggest Relationships for key-based joins, and Semantic Modeling (Beta, formerly Data Pro) for generating and refining whole models through conversation.
Understand how generative AI capabilities support the creation of Tableau Next components.
~8 min read · 19 practice questions
What the exam expects
- Pick the generative AI tool that produces the component a modeler needs: a calculated field, a description, a relationship or a whole model.
- Write a Draft with Einstein request and explain what happens to the generated calculated field after saving.
- Apply AI Semantic Description Generator (Beta) to objects, fields and calculated fields, including its prerequisites and fallback behaviour.
- Judge which fields and objects Suggest Relationships can use, and act on a recommendation with Review, Apply or Reject.
- Explain what Semantic Modeling (Beta) generates, its scope limits, and the controls in its conversational and single-prompt workflows.
Key terms:
- Draft with Einstein: the AI option in the New Calculated Field window that drafts a calculated field from a plain-language description.
- AI Semantic Description Generator (Beta): the Generate button in a description field that drafts metadata text for agents.
- Suggest Relationships: AI recommendations for relationships between objects in the Semantic Model Builder.
- Semantic Modeling (Beta): the agent that generates or refines a semantic model from natural language; older pages call it Data Pro or Semantic Modeler.
Key facts
Draft with Einstein prerequisites
- Detail
- Data Pro - Semantic Authoring AI in Data 360 Feature Manager plus Turn on Einstein
Saved Einstein calculated field
- Detail
- Added to the semantic model; usable to create relationships
Calculated field description
- Detail
- Generate works only on a valid calculated field
Standard object, no custom description
- Detail
- Agents use the existing standard description
Suggest Relationships field types
- Detail
- Numbers and text up to 50 characters; primary and foreign keys only
Unsupported objects for suggestions
- Detail
- Logical View and Shared Dimension data objects
Reject
- Detail
- That specific recommendation is no longer shown
Semantic Modeling scope
- Detail
- Objects, relationships, calculated fields; not metrics, groups, logical views
Recommendation scan limit
- Detail
- Up to 100 tables across the user's workspaces
Ignore in conversation
- Detail
- Cancels generation or the refinement request
Sources for this section (7)
- Create a Calculated Field with Einsteinhelp.salesforce.com (opens in a new tab)
- Calculated Field Descriptions (Beta)help.salesforce.com (opens in a new tab)
- Data Object Descriptions (Beta)help.salesforce.com (opens in a new tab)
- Suggest Relationshipshelp.salesforce.com (opens in a new tab)
- Semantic Modeling (Beta)help.salesforce.com (opens in a new tab)
- Create with Semantic Modeler (Beta)help.salesforce.com (opens in a new tab)
- Refine with Semantic Modeling (Beta)help.salesforce.com (opens in a new tab)
- What does Suggest Relationships base its recommendations on?
- When you finish editing an AI-drafted description on a data object field, which button saves it?
- Why might relationship suggestions appear in a model where nobody clicked a separate setting for them?
- When an admin wants to turn on Semantics AI in Tableau Next, which permission sets allow it?
The questions are free to read. Answers with their sources and self-grading are included with the pack.
Keep reading with the Tableau Next Consultant pack
- Core concepts
- Exam traps
- Worked scenario
The pack includes:
- All 21 lessons of the Tableau Next Consultant guide, with reading progress
- Interactive decision walks and step-by-step sequence steppers
- Practice exams purchase adds inline bank questions and topic practice (19 questions for this lesson)
- An official source cited for every claim
- An AI tutor that answers from the guide
- Choose guide-only, practice-only, or save $4 with both