Snowflake launches Cortex Code, an AI agent designed to automate data engineering and analytics, aiming to speed enterprise AI development with secure, context-aware tools.
Snowflake has unveiled Cortex Code, a new AI-powered coding agent designed to help organizations build data and AI applications faster and more securely.
Announced in London at the company’s BUILD 2026 conference, Cortex Code is positioned as a productivity tool for data teams—one that brings context-aware coding assistance directly into Snowflake and local development environments.
The company says the agent can translate complex data engineering, analytics, and machine-learning tasks into simple natural-language commands, enabling both technical and nontechnical users to move projects from concept to production more quickly.
“AI must become part of the systems teams rely on every day, not just an experiment,” said Christian Kleinerman, Snowflake’s executive vice president of product. “Cortex Code embeds AI directly into the development lifecycle with the data context and governance enterprises require.”
A Coding Assistant Built for Enterprise Data
Unlike generic AI assistants, Cortex Code is designed specifically for Snowflake’s ecosystem. It understands an organization’s data structures, governance rules, and compute environment, enabling it to generate code and workflows that align with existing controls.
Snowflake says the tool is already being used by customers, including Braze, dentsu, LendingTree, TextNow, United Rentals and WHOOP to automate tasks ranging from data pipelines to analytics and AI agent creation.
The agent is available through multiple interfaces. Users can work within Snowflake’s web interface via Cortex Code in Snowsight—expected to be generally available soon—or use Cortex Code CLI, now generally available, inside preferred tools such as Visual Studio Code or Cursor.
The goal, Snowflake argues, is to reduce the gap between experimentation and real-world deployment, a challenge that continues to slow enterprise AI initiatives.
Also Read: The AI Economy Speaks: Takeaways from Cisco’s AI Summit
New Tools to Support AI Workflows
Alongside Cortex Code, Snowflake announced several integrations intended to streamline AI development.
A forthcoming connection with v0 by Vercel will allow teams to build AI-powered data applications using natural language and deploy them securely within Snowflake. An integration with Brave Search, currently in public preview, will add real-time web knowledge to Snowflake’s AI tools, helping reduce inaccuracies and improve context for agents.
The company also introduced enhancements to its collaborative Workspaces environment, including shared notebooks and OpenID Connect authentication, to make team-based AI development more secure and seamless.
Early Customer Feedback
Several early adopters described Cortex Code as a meaningful boost to productivity.
“At dentsu, speed and governance must go hand in hand,” said Joe Tobey, head of data products engineering. “Cortex Code CLI lets our teams translate requirements into AI-powered solutions faster without disrupting established workflows.”
Executives at LendingTree, TextNow and Shelter Mutual Insurance echoed similar themes, emphasizing that the tool helps teams iterate more quickly while maintaining the controls required in regulated environments.
Also Read: Why AI Detection and Response is the New Baseline for Survival
A Broader Push Into Enterprise AI
Cortex Code forms part of Snowflake’s broader Cortex AI product suite, which aims to embed artificial intelligence across the full data lifecycle—from ingestion and transformation to analytics and application development.
For Snowflake, the strategy is clear: make AI development less specialized, more accessible, and tightly integrated with enterprise data.
Whether Cortex Code becomes a standard tool for data teams will depend on how well it delivers on those promises. But with organizations under pressure to turn AI ideas into tangible results, Snowflake is betting that an AI agent purpose-built for enterprise data will find a ready audience.
