Caleb Baechtold, AI/ML Principal Architect at Snowflake, provides an end-to-end walkthrough of how to build an application that can enable business users to ask questions in natural language against a curated set of data in Snowflake using Cortex Analyst.
Cortex Analyst is a fully managed talk-to-your-data (TTYD) service that leverages an agentic AI setup powered by a collection of state-of-the-art LLMs to reliably answer users' data questions. It is available as a convenient REST API that can be integrated into any application, while ensuring high security and governance through Snowflake access controls.
Follow along using this quickstart: https://quickstarts.snowflake.com/gui...
Learn more about Cortex AI: https://www.snowflake.com/cortex
00:00 Intro
03:00 Set up Snowflake account
04:19 Ingest demo dataset
07:35 Create Streamlit app for local testing
11:24 Test conversational analytics Streamlit app
14:59 Semantic model deep dive
21:29 Adding verified queries
26:00 Wrap up and conclusion
To learn even more, enroll in Snowflake's free Coursera course: https://www.coursera.org/learn/data-e...
Subscribe for more! http://www.snowflake.com/YTsubscribe/
Explore sample code, download tools, and connect with peers: https://developers.snowflake.com/