The Snowflake Intelligence platform lets business users build data agents that analyze data. The idea is that they can make better decisions based on that.
During its presentation at the developer conference BUILD 2024, Snowflake promises an easy way to create these data agents. This should give departments such as sales, marketing, and HR faster access to data. The data agents can pull data from the AI Data Cloud and third-party tools, such as sales transactions from a separate database, documents from SharePoint, and information from Google Workspace and Salesforce. Users can ask questions in natural language to the agent, as is common with chatbots.
““With Snowflake Intelligence, teams can easily create data agents that allow business users to talk to their enterprise data using natural language, and then analyze, summarize, and take action on those results from one unified platform,” said Head of AI Baris Gultekin.
Engine and governance
Snowflake Intelligence uses the managed Snowflake service Cortex AI as its engine. The Cortex Search component enables queries on unstructured data, while Cortex Analyst is designed for structured data. Updates to Cortex AI allow it to fine-tune Retrieval-Augmented Generation (RAG) and text-to-SQL, allowing data teams to process unstructured data and developers to build AI apps.
In addition, the tool features the Snowflake Horizon Catalog, which is compatible with open formats such as Apache Iceberg and Apache Polaris. This helps meet compliance, security, privacy, and discovery requirements.