Ground search on unstructured and structured data in Data 360 to use generative AI, improve analytics, and use automation tools.
- Rebuild Search Indexes Without Downtime
Rebuild search indexes in Data 360 without service interruptions. When you initiate an index rebuild, Data 360 creates a version of the search index in the background while users continue to query the current search index. After the rebuild is complete, the system automatically switches to the new index with no downtime. Your users experience consistent, accurate search results throughout the process. - Preview and Validate Citations in Retriever Playground
Test retriever citations in the Data 360 retriever playground to ensure accurate retrieval augmented generation (RAG) responses. Configure standard and custom citations for retrievers, select test data, and verify that citation links for retrieved chunks come from expected sources. Debug misconfigured citation setup or broken links in the playground before deploying retrievers in AI systems. - Create, Manage, and Test Retrievers in Data Cloud One Configurations
Use Data 360 AI Models to build and evaluate retrievers in Data Cloud One companion orgs. You can also view and use retrievers created in Data Cloud One home orgs. Previously, the AI Models option wasn’t available in Data Cloud One companion orgs. - Use Search Indexes and Intelligent Context with Platform Encryption
Create search indexes and intelligent context when Data 360 platform encryption is turned on. Previously, you couldn't use these features if platform encryption was turned on. - Fine-Tune, Test, and Optimize an Ensemble Retriever
Customize, test, and validate ensemble retriever responses in the Data 360 Retriever Playground. Avoid irrelevant output, eroded user trust, and increased costs by verifying response accuracy directly in the playground without calling an LLM or manually testing in local environments.