Snowflake Unveils AI Tools to Build Enterprise-Grade Agentic Apps

Snowflake has introduced a powerful suite of AI tools designed to help organizations build and deploy enterprise-grade agentic applications at scale. These platform enhancements focus on delivering trusted, accurate insights while maintaining robust security and governance. Key innovations now available include Snowflake Intelligence, an enterprise agent that enables users to ask complex questions in natural language, alongside significant upgrades to Snowflake Openflow and the Snowflake Horizon Catalog. These tools allow companies to unify structured, unstructured, and semi-structured data from diverse sources and catalogs, fueling sophisticated agentic AI capabilities. All of this operates within Snowflake’s secure and governed environment, which is built to be interoperable and avoids vendor lock-in.

A new collection of AI-native and collaboration tools has also been announced, empowering developers to construct, test, and launch enterprise-ready AI applications more quickly and safely. These improvements are designed to lower overhead and reduce the total cost of ownership, all within a single, governed platform.

Christian Kleinerman, EVP of Product at Snowflake, emphasized the company’s evolution, stating, “For over ten years, Snowflake has been a foundational element in the data strategies of global enterprises. Our next step is to infuse AI directly into that data, enabling every customer to uncover intelligence that is distinctly their own. The latest platform enhancements make this vision a reality, democratizing AI so that every employee can make faster, smarter decisions. This will fundamentally reshape how our customers drive innovation in the coming years.”

Snowflake Intelligence: A Trusted Enterprise Agent for All Organizational Knowledge

Snowflake Intelligence is now generally available to Snowflake’s global customer base, which includes more than 12,000 organizations. By asking a single question, users can conduct deep research and receive suggested solutions to business challenges that were previously difficult and time-consuming. This allows teams to progress from simply understanding “what” happened to grasping the crucial “why” behind the data. Engineered for reliability and scale, the agent is enterprise-ready with trust, governance, and security as foundational principles. Employees can confidently access and analyze data using natural language, while measures are in place to keep confidential information protected. This approach reduces uncertainty in decision-making and helps cultivate a pervasive data culture across the enterprise.

In just the last quarter, more than 1,000 Snowflake customers, including Cisco, Toyota Motor Europe, TS Imagine, and the USA Bobsled/Skeleton Team, have used Snowflake Intelligence to rapidly deploy over 15,000 AI agents throughout their operations.

Thierry Martin, Head of Data and AI at Toyota Motor Europe, reported, “Snowflake Intelligence has dramatically accelerated our development process, cutting agent deployment from months down to weeks. This has allowed our team to shift focus from writing code to concentrating on what really adds value: building rich business context and robust semantic models. The outcome is a notable competitive edge, we are delivering secure, compliant data solutions to market more rapidly, all while eliminating the risks associated with moving data.”

Leveraging industry-leading AI models from providers such as Anthropic, Snowflake Intelligence converts complex queries into conversational insights, effectively democratizing access to data and AI across the organization. Innovations from Snowflake’s AI Research Team have made Snowflake Intelligence up to three times faster on text-to-SQL queries, providing real-time answers without sacrificing accuracy. To further enhance the trustworthiness of responses, the team developed a novel evaluation approach called the Agent GPA (Goal, Plan, Action) framework. This method catches up to 95 percent of errors in standard dataset tests, approaching human-level error detection.

Enhanced Open Data Access and Flexibility for Agentic AI in the Enterprise Lakehouse

Snowflake has also rolled out advancements to Snowflake Horizon Catalog and Snowflake Openflow, both now generally available. These updates help enterprises seamlessly connect data from disparate sources and catalogs, enabling more robust, accurate, and trusted AI-driven insights. New features in Horizon Catalog provide essential context for AI and deliver a unified security and governance framework that protects and connects data across every region, cloud, and format, all while remaining interoperable and free from vendor lock-in. Meanwhile, Openflow allows enterprise users to securely automate data integration and ingestion from virtually any source, simplifying the process of keeping data centralized within the enterprise lakehouse.

Additional innovations are making data more accessible to AI agents like Snowflake Intelligence, all within a unified framework of security and governance.

By integrating open APIs from Apache Polaris (Incubating) and the Apache Iceberg REST Catalog directly into Horizon Catalog, Snowflake now delivers an enterprise-grade lakehouse that consolidates governance, security, and interoperable access control for data stored in open table formats.

New Developer Tools to Accelerate Enterprise AI Development

Snowflake has also rolled out a suite of developer tools designed to help organizations build, test, and deploy advanced, production-ready AI applications faster and with stronger security safeguards.

(Source: ITWire Australia)