Treasure Data unveils AI-powered command line tool

Treasure Data has launched Treasure Code, a command-line interface that uses agentic AI to handle day-to-day operations across its customer data platform.

The product is now generally available to Treasure Data customers. It is aimed at teams that manage data workflows, CDP configuration, customer journeys, and related automation. Treasure Data positions it as a way to cut manual steps and standardise operations through an interface that treats the platform "as code".

Treasure Code runs alongside the Treasure Data Intelligent Customer Data Platform and is aimed at data engineers, platform teams, marketing operations staff, and advanced users who want programmatic control over data operations and customer workflows.

The interface also works with AI agents that can run tasks within the platform, and includes integration with Claude Code for natural-language-driven creation and iteration.

Command line focus

Treasure Code is positioned as an AI-native command-line interface that lets users complete CDP tasks using natural language. Treasure Data frames it as an alternative to writing SQL or learning command-line syntax for common operational work.

The interface covers deployment workflows and configuration management. It also extends into marketing operations within the CDP, including work on segments and journeys.

Teams can manage CDP configuration as version-controlled code, supporting peer review and rollback processes that mirror software engineering practices.

Treasure Code is also designed to consolidate work that often spans multiple consoles and scripts, with a single command layer that automates deployments from development to production.

Operational claims

Treasure Data links the launch to rising operational complexity in CDPs and data platforms, arguing that manual processes slow iteration and increase operational risk as deployments and data volumes scale.

It also positions Treasure Code around AI-assisted iteration with human verification, saying the model improves operator productivity and reduces the time needed to build and change workflows inside the CDP.

Rafa Flores, Chief Product Officer at Treasure Data, said the company has seen strong early use.

"We've never seen such organic adoption from an AI-native product at Treasure Data, with over a quarter of our customers embracing Treasure Code in a matter of days."

Flores said customers often manage CDP deployments with very large datasets.

"Our customers deal with incredible complexity and scale operating a CDP, often with upwards of hundreds of millions of profiles and trillions of data points processed. Treasure Code reduces the operational burden of data operations, enabling brands to manage a CDP with fewer resources and free teams up for strategic, high-impact work," Flores said.

Early adoption

About one-quarter of Treasure Data's customer base is already using Treasure Code for technical operations, though the company did not provide a total customer count.

Tomohiko Sugiura, Executive Vice President at Dentsu Digital, said the natural-language interface has sped up the production of artefacts used in live work.

"With Treasure Code, it feels like we've added a legion of data engineers to the team. I can describe what I want in plain language and get production-ready SQL, segments, and workflows in minutes," Sugiura said.

Hideo Ota, Senior Vice President & Executive Officer at NTT Integration, said the product changes the skills required to build pipelines and workflows.

"Treasure Code is a transformative feature that fundamentally changes how we implement and operate our CDP. Historically, constructing data pipelines and workflows required highly specialized technical skills. With Treasure Code, however, we can now execute these tasks rapidly and easily using natural language. This allows even non-experts to maintain high-quality implementations, and we are eager to actively leverage this capability to accelerate our business," Ota said.

Treasure Data said Treasure Code is available immediately to existing customers through their account teams.