Data 360 as the foundation for AI
Why agents and models need unified, clean, real-time and governed data, how Data 360 grounds generative AI through RAG and search indexes, what the Einstein Trust Layer stores in Data 360, and the three ways AI Models (formerly Einstein Studio) delivers predictions.
Articulate Data 360's foundational role in generative and predictive AI.
~7 min read · 11 practice questions ·
What the exam expects
- Explain why agents and models need unified, cleansed, real-time and governed data, and match an agent's weakness to the Data 360 capability that fixes it.
- Describe how Data 360 grounds generative AI with retrieval augmented generation (RAG) over search indexes.
- Name the Einstein Trust Layer protections and the audit and feedback data Data 360 stores.
- Choose between building a model in AI Models, bring your own model (BYOM) and bring your own large language model (BYOLLM).
- Position Zero Copy federation as a way to feed agents and models without copying data.
Key terms: grounding adds business or customer context to a prompt; RAG retrieves relevant content, augments the prompt with it, then generates; search index stores chunked, vectorized content from a DMO; AI Models (formerly Einstein Studio) is Data 360's home base for predictive and generative models; audit trail is the prompts, responses and trust signals logged in Data 360.
Key facts
- Detail
- Unified profiles, transforms, real-time, zero copy, RAG, guardrails and analytics
- Detail
- Web or mobile app data stream or Ingestion API, in milliseconds
- Detail
- Real-time data graphs are required to use other real-time capabilities
- Detail
- Chunks in a CDMO, vector embeddings in an IDMO
- Detail
- Third-party LLMs retain no data and use none for training
- Detail
- Prompts, responses, trust signals and toxicity scores logged in Data 360
- Detail
- Turn on generative AI data collection and storage; install the report package
- Detail
- Create from scratch, BYOM (SageMaker, Vertex AI, Databricks), BYOLLM
- Detail
- Integrated in Data 360, so it sits under the Einstein Trust Layer
Sources for this section (6)
- Data 360 in Flows and Data 360-Powered Agentstrailhead.salesforce.com (opens in a new tab)
- Explore Real-Time Features in Data 360trailhead.salesforce.com (opens in a new tab)
- Use Search for AI, Automation, and Analyticshelp.salesforce.com (opens in a new tab)
- Einstein Trust Layerhelp.salesforce.com (opens in a new tab)
- Audit Trailhelp.salesforce.com (opens in a new tab)
- Analyze Data and Build Predictive Modelstrailhead.salesforce.com (opens in a new tab)
- Why can't BYOLLM connect a Databricks table to Data 360?
- When should a consultant recommend real-time ingestion for an agent?
- Where are Trust Layer toxicity scores stored?
- Why does AI Models sit under the Einstein Trust Layer?
The questions are free to read. Answers with their sources and self-grading are included with the pack.
Keep reading with the Data 360 Consultant pack
- Core concepts
- Exam traps
- Worked scenario
The pack includes:
- All 22 lessons of the Data 360 Consultant guide, with reading progress
- Interactive decision walks and step-by-step sequence steppers
- Inline practice questions and one-click topic practice (11 questions for this lesson)
- An official source cited for every claim
- An AI tutor that answers from the guide
- Unlimited timed practice exams