Snowflake unveils new tools for secure enterprise AI development

Snowflake has announced a set of new developer tools and platform updates aimed at enhancing enterprise-level agentic AI application development, focusing on improved workflow efficiency, security, and integration with open-source tools.

Productivity and security

The company detailed enhancements to its development environment, promising organisations increased productivity and a reduction in operational overhead. This comes as businesses continue to push for faster deployment and scaling of artificial intelligence (AI) solutions across large data environments.

Christian Kleinerman, Executive Vice President of Product at Snowflake, said,

"The success of enterprise AI hinges on having the most trustworthy data, and the most productive developers. By delivering a single, intelligent, and governed environment, we're not just accelerating code development and execution - we're giving every developer a shorter, simpler path to build enterprise-ready AI apps that actually drive value. This is the new blueprint for enterprise innovation and a demonstration of how Snowflake is delivering on its promise of limitless interoperability."

Collaboration and open source tools

Snowflake's new developer capabilities include enhanced collaboration features through its Workspaces environment, now generally available. Workspaces is designed to centralise development, allowing the creation and management of code across multiple file types. Integration with Git and Visual Studio Code (VS Code) offers version control and access to preferred integrated development environments without leaving the Snowflake platform.

The company has also expanded support for open-source solutions. With the availability of dbt Projects on Snowflake, organisations can deploy and manage data transformation pipelines directly within Snowflake. Andre Byfield, Principal Data Architect at Enlyte, commented,

"Snowflake's new developer capabilities have been transformative, empowering us to build data pipelines with the flexibility and interoperability we need, all while using the tools that best fit our workflow. dbt Projects on Snowflake allowed us to deploy and orchestrate our dbt pipelines directly on the Snowflake platform rather than having to build out that cloud infrastructure ourselves. This represented real cost and time savings for our lean data engineering team and delivered real-world value to our stakeholders."

Agentic AI and automation

Industry research suggests agentic AI is gaining momentum, with 20% of organisations already deploying AI agents and a further 54% intending to do so within the next year. Snowflake's updates aim to support these trends, introducing AI-native developer tools including Cortex Code, an AI assistant that enables natural language interaction with the Snowflake environment for optimising queries and managing resources.

Other updates include the release of Snowflake Cortex AISQL and Dynamic Tables, which streamline the creation of scalable AI-inference pipelines using SQL. The forthcoming AI Redact tool will enable the redaction of sensitive information from unstructured data, supporting privacy requirements when preparing multimodal datasets for AI applications.

Third-party integrations and performance

Snowflake highlighted interoperability as a priority, citing compatibility with major third-party developer products. The newly available Snowpark Connect for Apache Spark allows organisations such as VideoAmp to run existing Apache Spark code on Snowflake, with companies reporting significant performance and cost improvements. Customer usage data indicate a 5.6-times increase in performance and a 41% reduction in costs compared to other managed Spark environments.

The variety of supported tools enables engineers to focus on delivering analytics and business intelligence rather than maintaining infrastructure, with client organisations including InterWorks, Enlyte, NTT DOCOMO, and STARS using these features to streamline operations.

Chris Androsoff, Director of Data at STARS, commented,

"As a non-profit that delivers life-saving care every day, every dollar counts. When we rebuilt our data and analytics platform, we needed right-size tooling that balances capability with simplicity and cost. The moment dbt became part of the Snowflake ecosystem, the path was clear. Today we experiment, codify, test, deploy, schedule, and monitor our entire dbt workflow natively inside Snowflake. Consolidating on one platform has created helpful simplicity, improved cost transparency, and freed our engineers to focus on delivering value faster."

Data quality and code security

In an effort to ensure the reliability and security of data used in AI-driven applications, Snowflake introduced upgrades to its Data Quality User Experience and code security features. The Data Quality UI, now in public preview, enables automatic data reliability assessment and generates summaries for analysis. Code security updates offer new protection against unauthorised code access and potential data poisoning or model tampering.