July 2026 Summaries

The decision to retire the dbt Snowflake Native App from the Snowflake Marketplace is driven by the evolving landscape of dbt and Snowflake integrations, with a focus on more robust and capable options now available, such as dbt's Semantic Layer and Snowflake Intelligence. The app will enter a maintenance phase in July 2026 and be fully removed by November 2026, affecting only a small number of users who will receive direct support for transition. The original aim of integrating dbt-powered AI capabilities directly into Snowflake has been surpassed by advancements in AI and integration features, prompting dbt Labs to reallocate resources towards more valuable and widely adopted solutions like Semantic Views and native dbt execution on Snowflake. Meanwhile, dbt will continue to support Snowflake data teams with interoperable products and integrations, aiding in building reliable, governed data products for analytics and AI on Snowflake.

The dbt Summit 2026, taking place from September 15-18 at The Cosmopolitan in Las Vegas, is set to offer over 100 sessions designed to help analytics engineers, data team leaders, and business executives adapt to the evolving landscape of data and AI. The summit underscores the shifting role of data teams from traditional analytics to strategic drivers, as AI transforms the way data is consumed, emphasizing the need for a solid data foundation. Highlights include hands-on labs, peer exchanges, and breakout sessions focusing on governance, scaling, and AI integration. Key discussions will explore the practical application of AI tools, governance strategies for safe AI adoption, and career development in the data field. The event promises valuable insights for attendees aiming to leverage dbt for scalable, trusted data infrastructure and to empower stakeholders in the age of AI, with the opportunity to gain training and certification.

The collaboration between Fivetran and dbt Labs aims to redefine the data landscape with the introduction of dbt Core v2.0, a Rust-based engine under an Apache 2.0 license, enhancing the speed and efficiency of data transformation. This new version, along with features like dbt State and dbt Wizard, is designed to cater to the evolving needs of data infrastructure in the AI-driven era, where agents are significant consumers of data. dbt Core v2.0 promises faster execution, improved metadata handling, and extensive caching capabilities that can substantially reduce compute costs. The integration with Fivetran seeks to create an open data infrastructure that supports both human and agent-driven data consumption, emphasizing trust, scalability, and cost efficiency. The dbt State caching layer and the dbt Wizard coding agent further enhance development workflows and cost savings. As dbt continues to support open-source principles, it positions itself as a critical player in modernizing data stacks for the next decade, maintaining its commitment to innovation and community engagement.

The Open Semantic Interchange (OSI) project has transitioned to the Apache Incubator and is now named Apache Ossie (Incubating), reflecting a change in governance and long-term direction while maintaining its original mission and specification. This semantic layer and ontology open specification aims to standardize the expression of business metrics and concepts across various platforms, ensuring consistent interpretation and resolution of business definitions like "Monthly Active Users" across CRM, data warehouses, and BI tools. The transition to the Apache Software Foundation (ASF) is intended to preserve its status as an open standard under a vendor-neutral environment, supported by a diverse community of contributors from companies such as Snowflake, Salesforce, and dbt Labs. The project operates with an open development process, including public mailing lists, GitHub-based development, and a formal process for specification changes. As it grows under ASF governance, the community is focused on expanding the specification's expressiveness and building converters to facilitate broader adoption without disrupting existing systems.

The demand for data-driven insights is surging due to advancements in AI, yet many data teams face stagnant budgets, requiring them to optimize existing resources. dbt, a data control platform, offers a solution by streamlining data workflows through modular, reusable, and automated processes, significantly improving productivity and reducing maintenance burdens. According to an IDC report, companies using dbt have reported substantial gains, effectively recouping the equivalent of 58.7 full-time employees' worth of capacity across various data functions, including analytics, development, and data governance. By facilitating faster report delivery, accelerating development cycles, and improving onboarding times, dbt enhances team collaboration and reduces the incidence of data quality issues. These efficiencies are achieved through features like automated testing, documentation, and a structured deployment process, which help teams to focus on strategic initiatives rather than repetitive maintenance tasks. Overall, dbt empowers organizations to scale their data infrastructure effectively, ensuring long-term productivity gains without the need for additional hiring.

Integral Ad Science (IAS) addressed the "BI why" problem in business intelligence (BI) dashboards, where AI agents struggle to provide context and reasoning behind data metrics, by using Model Context Protocol (MCP) servers to connect AI chatbot agents to dbt and Databricks. At the Databricks Data + AI Summit 2026, Mars Dauer, IAS's Senior Director of Enterprise Data and AI, explained how his team utilized MCP as a universal adapter to quickly integrate new tools without custom API clients, thus enabling AI agents to access and interpret data lineage, validate data, and perform impact analysis effectively. This architecture involved embedding the agent in Looker to streamline context-sharing and employing multiple specialized sub-agents to efficiently handle different tasks, using varied models tailored to specific requirements. The approach has significantly reduced the time analysts spend resolving data issues, showcasing the potential for further enhancements in AI-driven data analysis and management.

Stephen Thibeault's guide addresses the disparity between AI adoption in coding and data pipeline management, as highlighted in the 2026 State of Analytics Engineering report, noting that 72% of teams prioritize AI for coding while only 24% do so for pipeline management. The guide explores why AI pipeline management lags, emphasizing that while AI-assisted coding is often an individual task, pipeline management requires team collaboration and alignment, making it more complex. Thibeault suggests layering AI onto existing ELT data architectures without overhauling them, focusing on high-value areas like code reviews, error triage, and ticketing systems to integrate AI into workflows effectively. The guide encourages starting with low-stakes use cases, piloting AI implementations, and continuously evaluating and maintaining AI systems to ensure they remain reliable and effective. It also stresses the importance of organizational buy-in from leadership to integrate AI meaningfully into daily workflows, with realistic expectations about the pace of efficiency gains and the necessary work to make AI systems trustworthy.

Data platforms, traditionally designed to store and serve data, are evolving into intelligence platforms that focus on making meaning available, as articulated by Dustin Dorsey. While data platforms excelled at infrastructure challenges, they fell short in bridging the gap between data accessibility and informed decision-making, a gap that was historically filled by human judgment. As AI systems begin to assume roles in reasoning over data, the foundational infrastructure of data platforms—primarily focused on storage and accessibility—proves inadequate for supporting AI's interpretative needs. The transition to intelligence platforms is not merely a technological upgrade but a philosophical shift towards prioritizing the semantic layer, which involves intentional data models, canonical definitions, and semantic governance. This shift demands organizations to invest in and maintain a robust knowledge layer, which is essential for ensuring consistent and reliable AI outputs. dbt and phData play critical roles in facilitating this transition by providing the tools and frameworks necessary for encoding and enforcing meaning within the transformation layer, enabling organizations to operationalize the intelligence platform philosophy and move beyond traditional data platform constraints.