Snowflake built an internal semantic layer—a unified data interface that translates raw data into consistent business definitions—to power AI agents and business intelligence tools across the company.
In July 2025, more than 5,600 employees ran over 320,000 queries through this semantic layer.
The approach dramatically improved AI accuracy: in benchmark testing, adding semantic context increased text-to-SQL accuracy from 20% to over 90%, and AI agents using semantic views run faster and consume fewer tokens than those querying raw data directly.
What happened
Snowflake built an internal semantic layer—a standardized data interface that sits between raw data tables and downstream consumers (dashboards, AI agents, and analytics tools)—to ensure consistent business logic across the company. In July 2025 alone, over 400 internal users ran more than 5,400 queries through the product data science agent, and over 5,600 employees across all Snowflake teams ran over 320,000 queries using the broader semantic layer in sales, HR, support, and other areas.
Why it matters
Without shared context, the same question (e.g., "What is an active customer?") yields conflicting answers across different tables and metric definitions. A semantic layer eliminates this by providing one trusted source of truth. AI agents using semantic views see faster execution and lower token costs, and in AtScale's benchmark testing, adding semantic context increased text-to-SQL accuracy from 20% to more than 90% across 40 business questions.
What to watch
Snowflake is using native dbt integration for version control and CI/CD on semantic views, semantic view materializations to optimize performance, and Cortex Agents with custom routing instructions to automatically find the right context for queries. The company also uses Cortex Sense to help route questions to relevant datasets across the long tail of data.
Ask the AI about this article →
Snowflake's investment in a semantic layer addresses a persistent problem in data organizations: the gap between raw data and actionable insight. When petabytes of data exist across millions of sources without standardized definitions, even straightforward business questions generate conflicting answers. The problem compounds with AI agents, which operate at machine speed and scale—if a Cortex Agent reports one number while a legacy dashboard shows another, there is no clear source of truth. By curating an internal semantic layer, Snowflake created a single, governed interface that translates physical database schemas into business logic that every consumer (human and machine) can trust.
The performance and cost benefits reinforce this architectural choice. Semantic views materialize preaggregated data, allowing AI agents to skip the time-consuming steps of finding, understanding, and sampling multiple raw data sources. Agents also become more token-efficient when they can reference curated dimension and metric relationships instead of reasoning through raw table joins. The July 2025 usage numbers—over 320,000 queries across the company—show that once the semantic layer is in place and accessible, adoption spreads quickly across multiple teams and use cases, from product analytics to HR and support.
AI-summarized, only the topics you pick — one digest a day via Email, Slack, or Discord.
Free · takes 30 seconds · unsubscribe anytime
Ask AI anything about this article. Q&As are published on this page for other readers too.
Astromech, an AI startup co-founded by Ben Lamm and geneticist George Church, raised $20 million in funding le…
Alibaba is guiding its AI cloud revenue toward a US$10 billion run-rate in the next quarter, signaling that it…
Ode, a venture arm of Anthropic, has acquired Casper Studios to expand its enterprise artificial intelligence…
SK hynix, the world's largest supplier of high-bandwidth memory (HBM), has published a technical roadmap for c…
Alibaba Group's June quarter results show cloud AI revenue growing 45%, while capital expenditure (investment…
Anthropic launched Claude Academy on August 20, a free learning site that explains AI fundamentals and how to…
