Announcing the Integration of AtScale and Snowflake Cortex Analyst

Summary

• AtScale and Snowflake Cortex Analyst have joined forces to enable AI-driven natural language querying (NLQ) on governed semantic models.
• This integration ensures accurate, consistent SQL generation, aligning AI-driven queries with business definitions.
• AtScale acts as the governance layer, optimizing query performance while maintaining enterprise-grade governance and data integrity.

We are excited to announce a powerful new integration between AtScale and Snowflake Cortex Analyst, an LLM-powered feature that allows users to reliably answer business questions based on structured data. Combining AtScale’s semantic layer with Cortex Analyst’s advanced text-to-SQL capabilities enables business users to query data using natural language while ensuring accuracy, consistency, and performance across their Snowflake environments.

What is Snowflake Cortex Analyst?

Snowflake Cortex Analyst is an AI-driven solution that enables organizations to process and analyze data using natural language queries (NLQ). With Cortex Analyst, users can ask questions in plain English and get trusted answers in real time, eliminating the need to rely on analysts or sift through BI reports. This integration accelerates self-service analytics, reducing time-to-answer for critical business decisions. However, ensuring the accuracy and consistency of these AI-generated queries requires a governed, structured data foundation—this is where AtScale provides the trust and scalability needed for reliable, data-driven decision-making.

How AtScale Enhances Cortex Analysts AI-Powered Queries

AtScale’s Semantic Layer seamlessly integrates with Snowflake Cortex Analyst, bridging raw data and business users. This partnership enhances Cortex Analyst’s ability to accurately interpret and generate SQL queries by providing predefined business logic, metadata, and performance optimizations. 

AtScale strengthens Cortex’s NLQ capabilities with:

• Increased Accuracy: AtScale’s semantic models provide clearly defined business terms, reducing ambiguity in natural language queries. When a user asks, “What were last quarter’s total sales across all regions?”, Cortex leverages AtScale’s metadata to interpret the query correctly, ensuring that the SQL generated accurately reflects the business’s reporting standards. Using this system, a joint effort by Snowflake and AtScale achieved 100% overall accuracy against AtScale’s NLQ Benchmark!
  
• Consistent Business Metrics: With AtScale as the single source of truth, queries executed via Cortex yield the same results as those performed in BI tools like Tableau, Power BI, or Excel. This consistency eliminates reporting discrepancies and fosters trust in AI-generated insights.
• Optimized Query Performance: By having Cortex Analyst generate queries against AtScale, we reduce the complexity of the question the Cortex is trying to solve. Instead of having to try and write complex queries with aggregations and joins, it instead writes a simple selection query against an AtScale logical table. AtScale then leverages its Query Engine to convert these simple queries into optimized multi-table operations that use metrics and joins sourced from trusted business definitions. This system ensures that AtScale and Cortex can deliver fast, efficient answers without slowing down Snowflake’s performance.
• Seamless Data Interaction: AtScale translates its semantic models into Snowflake’s Cortex format, allowing existing AtScale customers to leverage their models for Cortex Analyst, as well as existing BI use cases.
• Enterprise-Grade AI Guardrails: With AtScale, organizations can implement robust governance measures to avoid decisions being impacted by  AI-driven hallucinations. Cortex will not make assumptions or guesses about undefined metrics—if a user question falls outside the semantic model’s scope, users receive a clear prompt to refine their request.

Why This Matters for Snowflake Users

Integrating AtScale and Cortex Analyst fundamentally changes how business users interact with Snowflake data by making the experience seamless and intuitive. Business leaders who previously had to wait days or weeks for analysts to generate reports can now receive immediate, reliable answers to their ad hoc questions, significantly accelerating decision-making processes. Traditionally, business users often struggle with SQL while data teams grapple with anticipating every possible query. By enabling natural language queries on a governed semantic layer, this integration eliminates the communication gap between these groups, fostering more efficient collaboration.

Additionally, any updates made to an AtScale semantic model are automatically translated for Cortex Analyst, ensuring that AI-generated queries consistently align with the latest business definitions without requiring manual intervention. This level of automation streamlines governance while reducing errors. Organizations can also leverage the combined power of Snowflake Cortex and AtScale as part of their modern AI/BI strategy, allowing insights to scale across teams while maintaining strong governance and control over data assets.

What’s Next?

With AtScale’s Semantic Layer powering Snowflake Cortex Analyst, enterprises can unlock the full potential of AI-driven analytics while ensuring their data remains accurate, consistent, and governed. This integration enables organizations to modernize their decision-making processes and harness AI-driven insights at scale.

Ready to experience the power of AtScale and Cortex Analyst? Discover how this integration can transform your data strategy and empower your teams to make faster, smarter decisions.