Transforming Analytics Through Innovation
New capabilities accelerate developer workflows, automatically reduce compute spend, and bring intelligent AI assistance to structured data
On the keynote stage at Coalesce, dbt Labs, renowned for setting the gold standard in structured data compliance, has proudly showcased groundbreaking enhancements within its dbt platform. With the introduction of features powered by the cutting-edge dbt Fusion engine, the company is optimizing cost and boosting performance. Additionally, dbt Labs unveiled dbt Agents, an array of intelligent AI tools developed to streamline operations, enhance governance, and provide trustworthy AI outcomes.
"Open standards and AI are propelling the next phase of analytics, and the dbt Fusion engine is the crucial link for data teams to embrace this future," stated Tristan Handy, the visionary founder and CEO of dbt Labs. "Fusion not only offers comprehensive context but also equips developers with advanced tools and error-correction mechanisms essential for both human and AI tasks. This is the key to crafting next-generation, AI-driven data infrastructure."
With Fusion, currently in Preview for select projects, teams leveraging top platforms such as BigQuery, Databricks, Snowflake, and Redshift are experiencing significant benefits. The innovative features of Fusion provide an unprecedented developer experience, leading to extensive optimization of compute expenses. Specifically, the new state-aware orchestration feature enables organizations to decrease compute costs by about 10% simply by activating it, ensuring that only models that have changed are processed, preventing unnecessary expenses without necessitating massive project rewrites. Furthermore, teams can fine-tune their pipelines by specifying data freshness requirements, with state-aware orchestration identifying the most efficient job execution path. This has the potential to deliver an additional 15% savings, with some early adopters reporting total savings exceeding 50%—an impactful reduction for organizations relying on data infrastructures.
However, Fusion’s capabilities extend beyond mere cost savings; it also supports diverse and evolving analytics applications. dbt-powered pipelines can now effectively create and manage Apache Iceberg tables within Snowflake and Databricks, simplifying the transition toward open table formats and enhancing cross-platform data portability. Developers can now take advantage of the new dbt VS Code Extension, available in Preview, allowing for local execution of Fusion, fostering tighter development cycles and ensuring alignment with production settings. Meanwhile, dbt Insights, now Generally Available, leverages Fusion's language server, consolidating definitions, lineage, cost, performance, and reliability into one dashboard for quicker and smarter decision-making.
James Dorodo, VP of Data Analytics at Bilt Rewards, emphasized the transformative impact of dbt Fusion by stating, "The advancements brought by the dbt Fusion engine significantly alleviate many challenges in our dbt development processes. We anticipate a dramatic acceleration in our operational velocity."
Enhancing Data Analytics with Agentic AI
As a vital actor in establishing standards for AI-ready structured data, dbt is enabling the introduction of governed AI agents that utilize dbt’s unique contextual power to expedite analytics processes while ensuring quality, trust, and governance. dbt Agents are seamlessly integrated into the platform and include:
- Developer Agent: This tool aids in explaining logic, flagging duplicates, validating data, and authoring or refactoring content from prompts in VS Code or dbt Studio, ensuring a swift, safe shipping process.
- Discovery Agent: This AI agent assists users in locating the appropriate datasets and definitions while pinpointing trusted sources to facilitate quick exploration.
- Observability Agent: By monitoring jobs, this agent can detect probable root causes and suggest corrections, thereby reducing the time spent on manual remediation.
- Analyst Agent: Embedded in dbt Insights, this tool responds to inquiries about models, jobs, and metrics, revolutionizing the speed of insight generation.
These innovative agents are designed to embed AI technology deeply into the Analytics Development Lifecycle. By enhancing quality and governance, they help teams achieve faster and better outcomes, ensuring that AI translates into real business advantages. The structured data context they provide is made universally accessible to AI systems via the remote dbt MCP server, now available to all. This cloud-based solution links AI tools to dbt projects without requiring local installations. Consequently, dbt's context, tooling, and error-correction features are now accessible for model providers and IDEs like OpenAI, enhancing the reliability of AI applications.
Øyvind Eraker, Senior Data Engineer at NBIM, noted, "dbt is our governance backbone. The MCP exposes that structured context to our chat experience, enabling every NBIM employee to ask questions and receive reliable answers. With dbt, Claude, and Snowflake driving our chat functionalities, our user adoption has soared by 10 times, while demand for core data team assistance has decreased significantly. AI solutions at our scale depend on prioritizing metadata, quality, and governance, and dbt establishes that essential foundation."
Commitment to Open Standards
In addition, dbt Labs made a significant announcement regarding MetricFlow, which has transitioned to being fully open source under an Apache 2.0 license. This move highlights dbt Labs' dedication to the Open Semantic Interchange initiative, promoting standardized metrics and semantics across tools to maintain consistent, trustworthy outcomes in analytics and AI workflows. Southard Jones, Chief Product Officer at Tableau, expressed excitement over the launch of dbt Agents, affirming the shared commitment among dbt, Tableau, and Salesforce to empower agents to benefit customers fully. This collaboration aims to establish consistent definitions for agents to drive meaningful business results.
For those interested in exploring these exciting product innovations introduced at Coalesce, registration is still open for virtual access to keynotes and selected sessions on the official dbt site.
About dbt Labs
Since its inception in 2016, dbt Labs has been dedicated to equipping data practitioners with the tools necessary to create and disseminate vital organizational knowledge. As the recognized standard for AI-ready structured data, dbt, enhanced by the dbt Fusion engine, provides the performance, context, and reliability organizations require to scale analytics effectively in the age of AI. Over 80,000 data teams around the globe, including industry leaders such as Siemens, Roche, and Condé Nast, rely on dbt to facilitate their data achievements.
Learn more at getdbt.com, and follow dbt Labs on LinkedIn, X, Instagram, and YouTube.
Frequently Asked Questions
What is the dbt Fusion engine?
The dbt Fusion engine is a powerful tool aimed at optimizing data processes, enhancing performance, and supporting new analytics use cases.
How does dbt Labs ensure cost savings?
Through its state-aware orchestration feature, dbt helps teams decrease compute costs significantly by ensuring that only necessary models are run.
What are dbt Agents?
dbt Agents are intelligent AI tools integrated into the dbt platform to assist with analytics, governance, and data quality enhancement.
Who can benefit from dbt's innovations?
Data practitioners and teams utilizing analytics platforms like BigQuery or Snowflake can leverage dbt's innovations to optimize workflows and achieve better results.
What commitment has dbt Labs made towards open source?
dbt Labs has committed to open sourcing MetricFlow to support the Open Semantic Interchange initiative, ensuring trusted and consistent analytics across various tools.
