Data and AI teams shouldn’t have to choose between speed and governance. Yet building semantic models manually remains one of the most time-intensive steps between raw warehouse data and trusted business insight.
In this demo, see how AtScale’s MCP Server acts as the bridge between AI agents and your semantic layer, enabling Claude, ChatGPT, or any custom agent to create models from warehouse metadata, validate them automatically, and serve consistent, governed query results back to any client.
In this demo, AtScale Co-Founder and AtScale Labs leader Dianne Wood demonstrates how AtScale's MCP server enables AI assistants like Claude, ChatGPT, and custom agents to automatically create, refine, deploy, and query enterprise semantic models.
What You’ll Learn
- How AI agents use the AtScale MCP Server to discover connections, inspect schemas, and generate complete semantic models from Snowflake
- How AtScale’s two-pass model generation builds intelligent hierarchies, semi-additive metrics, and time-series logic automatically
- How SML validation catches and corrects errors before deployment
- How to explore and query deployed models through natural language — with results that are consistent across AI clients
- Why governing AI at the semantic layer produces deterministic outputs that don’t vary by how a question is phrased
