Guided by the mission of simplifying data workflows for every builder, wherever they operate, Snowflake is unveiling a series of innovations that save developers from toggling between fragmented tools and windows and, instead, let them focus on doing their best work — pushing the boundaries of modern app development for an agentic AI future and beyond. From the recent introduction of Workspaces to the groundbreaking advances built into Cortex Code, our new AI coding agent announced today, Snowflake remains committed to the vision of a single, integrated environment with DevOps built in. 

Simplify data-native building with Cortex Code

Today, every data practitioner, regardless of role or technical expertise, is expected to build. And while they have a vast array of tools available to them, end-to-end data workflows remain complex and time intensive. But general-purpose tools are not data-native and lack seamless interactions with the data ecosystem, causing both builders and their organizations to struggle with the speed of innovation required to remain truly cutting-edge.

To help developers create code faster and more accurately, Snowflake is introducing Cortex Code, a Snowflake native AI coding agent designed to turn complex data engineering, analytics, ML and agent-building tasks into simple and informed interactions with high accuracy and trust — all in natural language. While third-party general purpose coding agents are powerful, they lack the native awareness of Snowflake metadata, catalog information and role-based access control required to deliver context-aware automated workflows. By contrast, Cortex Code is purpose built to accelerate the entire data lifecycle with enterprise-grade governance and reliability.

This deep platform intelligence enables a breadth of impact across the organization:

• Speeds time to production by enabling faster data engineering, advanced analytics and the development of agents and applications

• Empowers all users — from technical experts to nontechnical teams — to build with data confidently

• Simplifies complex tasks and enables sophisticated workflows while increasing productivity significantly

Built to interact seamlessly with the entire Snowflake ecosystem, Cortex Code works wherever builders operate — across Snowflake experiences and developer environments — adapting naturally to existing workflows instead of imposing new ones. Builders can use Cortex Code within the Snowflake platform through Cortex Code in Snowsight (generally available soon) or within their preferred terminal or code editor such as VS Code or Cursor with Cortex Code CLI (now generally available). 

A rich set of built-in and extensible skills provides expert Snowflake workflows, enabling design, implementation, optimization and operational automation end to end. This combination of platform intelligence and skill-based execution dramatically compresses the distance between idea and production, even for advanced use cases. With Cortex Code, Snowflake expertise becomes an always-on capability of the platform, empowering everyone to build faster and with confidence.

Build better apps with familiar tools

With AI, the barrier to building software has collapsed, but the barrier to shipping enterprise-grade apps remains high. Snowflake Apps breaks this cycle by fundamentally changing the geography of development. Now, almost anyone can turn ideas into apps in minutes using the open frameworks and ecosystems they already love. With Snowflake’s unified platform, developers can build enterprise-ready apps without the integration tax of fragmented infrastructure or brittle ETL pipelines.

Thanks to a new integration with Vercel v0, anyone can create apps just by describing them in natural language. These apps connect automatically to Snowflake data and run directly in any secure Snowflake account through Snowpark Container Services (SPCS). Because they are built where the data lives, these apps instantly inherit Snowflake’s native security and governance, allowing builders to bypass traditional security bottlenecks. Whether teams are building internal tools or distributing customer-facing solutions via the Snowflake Marketplace, Snowflake provides the foundation to go from a single prompt to a production-ready app.

For more seasoned developers, Snowflake allows users to run unstructured analytics on text and images with industry-leading LLMs through Cortex AI Functions. These functions, which can be called using SQL or Python syntax, allow developers to process and analyze multimodal data at scale by applying AI right where their data lives, using languages they are already fluent in.

Manage projects and collaborate effectively with DevOps baked in

Easy accessibility to cutting-edge AI tools is an integral part of Snowflake’s deep commitment to empowering developers. By offering world-class tooling and integrations with a wide range of third-party providers, Snowflake aims to give builders the luxury of choice without the fear of lock-in or costly complexity. 

That begins with Workspaces, a unified environment for developing end-to-end data projects — from writing SQL and Python code to managing a variety of project types such as dbt Projects on Snowflake and Snowflake Notebooks v2. Snowflake Notebooks v2 features a new underlying engine that provides Jupyter Notebooks compatibility, improved performance, Workspaces integration and advanced ML support. 

We now have shared Workspaces that allow teams to work together in a single environment, with each team member having access to the shared workspace, while maintaining robust security and access controls. Snowflake’s deep integration with Git also provides version control and seamless collaboration for all of your Snowflake objects using OpenID Connect (OIDC) to secure the entire infrastructure. Any Git-enabled platform can be used, including locally hosted solutions. Github Actions are fully supported and can be used to create a comprehensive continuous integration/continuous deployment solution. 

Git integration also lets developers use their favorite IDE to work on any aspect of Snowflake. VS Code integration means that a developer can do everything in their environment of choice and share it easily with the rest of their team. Snowflake CLI provides a comprehensive command line interface for building and working directly with your Snowflake objects. You can then use it to automate execution using tasks that can be executed individually or set to run on a schedule.

And because FinOps necessarily plays an important part of any development team’s charter, Snowflake provides tools to easily monitor Snowflake usage and understand how to increase efficiency and reduce costs. Using a combination of Cortex Code and Snowsight dashboards, you can query Snowflake to find out what your most used queries are and have Cortex Code suggest updates to make your implementation more efficient.

A world-class environment for all

Across the board, developers really want three things: to build with familiar tools; to manage projects in a governed way using formalized processes; and to simplify AI workflows to drive efficiency. 

Snowflake is committed to delivering on all three fronts. With an integrated developer environment with support for SQL, Python and Scala, along with various project types, Snowflake empowers builders to develop the way they want, and then connect with others using integration tools that allow for ideal DevOps and CI/CD implementations. Our growing suite of tools and capabilities aim to make building with AI — and for AI — easier and more efficient for all. 

To get started, explore these features further in the documentation below:

• Cortex Code
• Cortex Code CLI
• Workspaces
• Snowflake Notebooks
• Cortex AI Functions