Fivetran + dbt Labs finalise merger to build the data infrastructure for trusted AI agents
-
dbt Core v2.0 (alpha): The open sourcing of the dbt Fusion engine runtime, released as dbt Core v2.0 under an Apache 2.0 license, giving every practitioner the dbt experience they know on a faster, more capable foundation. Additionally, the locally installable distribution of dbt gives developers free access to the full breadth of Fusion’s capabilities – core language features and warehouse adapters – with the ability to seamlessly unlock additional platform features by logging in directly from the terminal.
-
dbt State (preview): dbt State acts as a caching layer for data pipelines. It only builds what’s changed and skips what hasn’t, helping companies reduce underlying infrastructure costs by 30 percent or more.
-
dbt Wizard (beta): dbt Wizard brings autonomous assistance for model authoring, refactoring, and debugging, grounded in full dbt project context, including lineage, tests, contracts, and defined metrics. The result is governed recommendations and trusted SQL generation that reflect how enterprise data is actually structured and defined.
-
Agents Schema: An open source standard for agentic context that designates a single schema in the warehouse or lake as the shared context layer for AI agents. Metric definitions, semantic models, dbt lineage, and business documentation are stored in plain SQL tables and can be published from existing systems through tools such as GitHub Actions, metadata connectors, or custom integrations. Compatible with any warehouse, lake, ingestion tool, or SQL-capable agent, Agents Schema gives organisations a customer-owned context layer that works within existing security and governance policies, improves token efficiency through richer context, and eliminates the need for new infrastructure or vendor-locked agent systems.
Customers building with Fivetran + dbt Labs
“With Fivetran and dbt, what used to take months now happens in weeks,” said Akshay Agrawal, Director of Data Engineering at Zendesk. “It gives the business faster access to trusted data and creates the foundation we need to scale analytics, agents, and AI across the enterprise.”
“Our focus now is on how we operationalise AI across Inova. With Fivetran and dbt, we’re creating the foundation for AI agents and applications that can act on trusted, governed data – not just generate insights, but drive action,” said Jon McManus, Chief Data and AI Officer, Inova Health.
“The combination of Fivetran and dbt isn’t just about efficiency today,” said Lakshmi Ramesh, VP of Data Services at Tinuiti. “It’s about being ready for what’s next. Analytics, AI, and agentic workflows all run on trusted data, and together, Fivetran and dbt are the data infrastructure that makes it all possible.”
“By building an AI-ready data foundation with Fivetran and dbt, we’re improving how teams across Shutterstock access and operationalise trusted, real-time data for analytics and emerging AI-driven workflows,” said Jitesh Kumar, Senior Software Development Manager at Shutterstock.
The combined innovations are now available and will be showcased throughout Snowflake Summit 2026. Summit attendees can visit the Fivetran booth (booth #2313) or the dbt Labs booth (booth #2112) to connect with product experts, experience demonstrations of the combined innovations, and explore customer use cases powering trusted AI agents at scale.