June 2026 Summaries

Preset has expanded its Managed Private Cloud (MPC) offering to Microsoft Azure, making it the only independent BI platform providing managed private cloud deployments on AWS, GCP, and Azure. This move eliminates the need for organizations in sectors like healthcare, financial services, and government, which require data residency and VPC-level isolation, to choose between managing Apache Superset™ themselves on Azure or switching cloud providers. MPC on Azure offers features such as automated deployments, security patches, elastic scalability, and compliance with SOC2 Type 2, PCI-DSS, and HIPAA, while also supporting multiple isolated Apache Superset™ workspaces. This addition increases Preset's deployment options to five, covering multi-tenant SaaS on AWS, Managed Private Cloud on AWS, GCP, and Azure, as well as customer-operated On-Prem solutions, contrasting with competitors like Power BI, Tableau, and Looker, which have more limited cloud deployment options.

In June 2026, the Preset Team released a newsletter that highlights the latest updates and features in Preset, encouraging users to explore its visual capabilities by offering a free trial. The newsletter also invites readers to subscribe to the blog for weekly digests of new posts, emphasizing the team's commitment to keeping users informed and engaged with the latest developments in their offerings.

Open source software development has been transformed by AI, which automates many tasks traditionally performed by humans, such as writing first drafts of features, reviewing code, and even merging pull requests, allowing projects to progress faster. However, this increased speed raises concerns about community health, as metrics like activity charts don't capture whether contributors feel engaged or whether diverse organizations are involved. Humans still play essential roles in fostering collaboration, communication, and community stewardship, which AI cannot replicate. These roles include welcoming newcomers, maintaining project tidiness, connecting contributors, nurturing ideas, and ensuring fair governance. While AI excels at repetitive tasks, the human element remains crucial in creating a welcoming environment that encourages sustained contribution and growth. The essence of community building is about making contributors feel valued and ensuring they have a supportive space to thrive, highlighting the irreplaceable value of human interaction and engagement in open-source projects.

In June 2026, Fivetran and dbt Labs merged to form a $4.4 billion entity, significantly impacting the data industry by controlling both data ingestion and transformation for over 100,000 teams. This consolidation has raised concerns about increased costs and reduced flexibility for data teams, as the merger shifts incentives towards bundled pricing and ecosystem lock-in, with some Fivetran customers experiencing substantial price hikes. Across the data analytics landscape, prices are rising, as seen with Salesforce's 9% increase in Tableau prices and other BI tools like ThoughtSpot and Looker maintaining high costs. The BI layer presents an opportunity for optimization due to its lower switching costs compared to the ingestion and transformation layers. Preset, built on the open-source Apache Superset, offers a competitive alternative with a focus on avoiding cloud vendor lock-in, providing extensive database connectivity, and leveraging community contributions. The text emphasizes the importance of considering total cost of ownership and the potential savings offered by Preset compared to traditional proprietary BI tools, especially in a market where data infrastructure is becoming increasingly bundled and expensive.

Preset's AI agent skills are designed to enhance the reliability and accuracy of agents working within production data platforms by providing them with targeted guidance for specific tasks. These skills help prevent agents from making errors by ensuring they use the correct APIs, follow proper workflows, and avoid destructive actions without user confirmation. The skills are akin to onboarding instructions, allowing agents to operate with a clear understanding of the environment's rules, thus minimizing risks such as deleted datasets or leaked credentials. Preset's approach involves benchmarking these skills against real tasks to demonstrate their effectiveness, revealing that skills significantly improve agent performance in 30 out of 31 cases. Although skills add some complexity by increasing the context needed for agent operations, this trade-off is viewed as beneficial due to the enhanced safety and consistency they bring to data operations. These skills are openly available for integration into existing systems, promising improved trust and efficiency in data management tasks.

Apache Superset 6.1 introduces significant advancements in extensibility and automation, featuring a matured Extensions framework, the new Model Context Protocol (MCP) service for AI integration, and the Global Task Framework for streamlined task management. With contributions from 136 developers, including 69 first-timers, the release includes over 1,500 pull requests that enhance AI tooling, charts, dashboards, database support, and more. The Extensions framework allows for more practical use through stable APIs, while the MCP service offers AI assistants a structured way to interact with Superset. The Global Task Framework unifies task tracking and management, and the introduction of Matrixify enhances chart layout capabilities. Additional improvements include auto-refreshing dashboards, dynamic currency formatting, and expanded database connectors. The release emphasizes community involvement, with opportunities for testing and contributing to future developments.

Preset has launched Preset Agent Skills, an open-source library designed to enhance AI agents' ability to interact with Preset, Apache Superset™, and related data environments by providing them with domain-specific expertise. These skills are not mere plugins but are comprehensive instruction sets that enable AI tools like Preset Chatbot, Claude Desktop, and GitHub Copilot to perform analytics tasks with a level of precision akin to that of a data engineer. The skills library consists of three packages: preset-api-skills, preset-mcp-skills, and preset-cli-skills, each covering different aspects of working with Preset, from API interactions to command-line interface operations. The integration of these skills allows AI agents to execute tasks like SQL execution and dashboard management without making erroneous assumptions, thereby increasing reliability and safety in production workflows. By being open-source and published under the Apache 2.0 license, Preset Agent Skills are accessible for customization and auditing, allowing organizations to ensure that their AI operations align with specific internal practices. This initiative aims to improve the accuracy and efficiency of AI-driven analytics by embedding tested workflows directly into AI behavior, ultimately enhancing both user experience and operational consistency across different AI tools.

May marked a period of significant advancements for Superset, with 66 contributors merging 475 pull requests, reflecting a focus on deepening the platform's capabilities following April's foundational developments. Notably, Superset's MCP service expanded to offer comprehensive tool support for AI assistants, while localization efforts introduced new translations and AI-assisted backfill systems. Semantic layers were revamped as extensions, and support for schema-less databases was integrated, enhancing dataset creation. The introduction of soft-delete infrastructure laid groundwork for future data management capabilities, and enhancements to charting, dashboards, and security measures were implemented. The community saw growth with 36 new contributors and 560 new GitHub stars, while 18 people made their first contributions, demonstrating increased engagement and collaboration within the Superset ecosystem.