FabCon + SQLCon 2027 content call opens with new dedicated tracks for Fabric platform teams

Source: fabriccon.com

The call for speakers for FabCon + SQLCon 2027 in Atlanta is now open, bringing a heavily restructured set of submission tracks tailored directly to enterprise data architectures. As organizations scale their Fabric adoption beyond initial pilots, the updated track structure highlights a clear shift toward platform engineering, operational intelligence, and enterprise-wide application integration. For data leadership, this signal reveals where Microsoft is focusing its engineering momentum and how enterprise platforms are expected to evolve over the next year.

Enterprise governance and developer workflows gain dedicated focus

The newly refined tracks separate Fabric administration from core developer experiences, reflecting how mid-market and enterprise organizations structure their internal platform teams. Submissions under “Fabric Platform: Admin & Governance” will focus heavily on automated data protection, OneLake catalog expansion, access controls, and cross-estate compliance.

Meanwhile, “Fabric Platform: Dev Experiences” prioritizes technical depth in CI/CD pipelines, API-driven extensibility, deployment automation, and capacity monitoring via the Metrics App. This explicit separation helps platform architects isolate operational guardrails from developer velocity, ensuring enterprise governance doesn’t stall delivery pipelines.

New operational and planning modules extend the Fabric core

A major shift in the 2027 agenda is the inclusion of specialized operational tracks such as “Fabric Apps / Rayfin” and “Fabric Planning”. The addition of application development frameworks – featuring AI-assisted TypeScript tooling, custom GraphQL APIs, and direct OneLake integration – signals a move toward building full-stack data products directly on top of Fabric.

Simultaneously, the dedicated planning track addresses enterprise writeback scenarios, financial forecasting, and variance analysis linked directly to governed semantic models. These specialized areas demonstrate that Fabric is expanding from a traditional analytical warehouse into an active operational engine.

Alignment strategies for enterprise platform leaders

For Chief Data Officers and Data Platform Managers evaluating these developments, the conference structure provides a roadmap for internal skill development and platform maturity. The emphasis on dedicated operational tracks – ranging from Real-Time Intelligence to Fabric IQ and Agent integrations – means enterprise architects should re-evaluate how their teams handle real-time data movement and conversational analytics over governed sources.

Organizations planning their 2027 roadmaps can leverage these core track definitions to align internal center-of-excellence standards with Microsoft’s primary engineering direction.

Learn more on Fabric Updates Blog

Microsoft Fabric upgrades data agent visual rendering engine for enterprise reporting

Source: Microsoft

Microsoft Fabric has upgraded its data agent conversational experience with native Fabric Visuals, replacing the previous standalone chart rendering engine. By embedding the core technology used across Fabric Apps, the data agent now delivers standardized, higher-fidelity visual responses natively within natural-language data chats. This change addresses a major gap in conversational analytics: generating visual outputs that match the design, precision, and usability standards expected in enterprise reporting.

Visual standardization across the platform

The migration to Fabric Visuals brings immediate improvements to visual fidelity, including sharper axis scaling, better legend placement, cleaner line markers, and refined hover tooltips. Supported chart types now include single and multi-series line charts, clustered bar charts, stacked area charts, scatter plots, and pie charts across all Fabric data sources, such as lakehouses, warehouses, and Power BI semantic models.

Because the agent uses the same rendering engine as the rest of the ecosystem, users no longer face inconsistent visual formatting when moving from chat-based analytics to broader reporting dashboards.

Lowering friction for self-service analytics

For enterprise architecture and data management teams, consistency in data presentation is crucial for trust and adoption. The update eliminates visual friction during conversational data exploration, allowing business leaders and analytics teams to validate trends, spot outliers, and digest complex metrics instantly without jumping into a full Power BI report.

Since visual generation activates automatically based on user intent or natural-language prompts like “show top customers” or “visualize revenue as a bar chart,” the feature speeds up decision-making without adding administrative overhead or extra configuration requirements.

Driving adoption through reliable design

By bringing professional visual standards to conversational AI interfaces, Microsoft reinforces the data agent’s role as a dependable tool for enterprise self-service analytics. Corporate data leaders can now encourage broader operational teams to interact with Fabric data platforms using natural language, confident that the resulting visual output remains clear, accurate, and aligned with company-wide reporting standards.

Learn more on Fabric Updates Blog

Microsoft Fabric data agents reach GA with intelligent routing to protect enterprise data governance

Source: Microsoft

The General Availability of data source routing in Microsoft Fabric data agents solves a critical enterprise governance bottleneck: stopping multi-source AI agents from delivering confident, well-formatted answers derived from the wrong data assets.

Smarter query orchestration eliminates silent hallucination risks

When a single conversational agent connects simultaneously to a lakehouse, warehouse, Power BI semantic model, and Eventhouse KQL database, picking the right underlying source is deceptively complex. A query like “how did checkout perform?” could mean revenue (lakehouse), system error rates (KQL logs), or a curated business metric (semantic model).

Fabric orchestrators now split this process into two distinct stages: a specialized router first locks onto the correct source by evaluating metadata, example queries, and intent before triggering language-specific query generators (NL2SQL, DAX, or KQL). This decoupled architecture prevents agents from generating technically correct SQL or DAX queries against completely inappropriate datasets.

Metadata filtering and custom intent rules scale large data estates

Enterprise schemas are often too massive for an LLM to evaluate entirely in a single context window without latency spikes. Fabric addresses this by evaluating metadata subsets for routine queries, while deploying a dedicated full-schema routing tool only when ambiguity arises. For platform teams, control is fully configurable.

Data architects can restrict schema visibility to relevant entities, define one-sentence intent descriptions, map domain-specific example queries, and enforce strict routing rules in the agent instructions (such as forcing logistics questions directly to the logistics lakehouse). This governance layer gives data teams deterministic control over how conversational AI navigates heterogeneous Fabric items.

Why data platform leaders should optimize routing strategy now

For CDOs and Data Platform Managers, unmanaged AI routing presents a hidden operational risk: plausible wrong answers that silently erode business trust faster than outright execution errors. By enabling fine-grained metadata pruning and custom intent routing, Fabric data agents allow enterprise teams to safely expose unified, multi-source data architectures to non-technical users.

Platform teams can audit routing decisions directly within run step logs, turning black-box LLM decisions into an inspectable, tuneable engineering workflow that scales securely across large data estates.

Learn more on Fabric Updates Blog

Power BI August update strengthens enterprise modeling and governance controls

Source: Microsoft

Microsoft has expanded Power BI’s enterprise capabilities this August, prioritizing governance least-privilege access, Direct Lake scalability, and local-to-cloud developer workflows. This release introduces critical schema-refresh controls for semantic models, expands Direct Lake hybrid capabilities, and removes security friction in embedded app distribution, giving platform leaders tighter control over data hygiene and operational costs.

Tightening workspace security with least-privilege model permissions

Microsoft Fabric apps now support Read-only permissions on underlying semantic models for business consumers, eliminating the requirement to grant broader Build access. Previously, distributing report apps forced platform managers to grant users Build rights, which opened untracked access to query endpoints, Excel exports, and downstream model creation.

By decoupling app access from model creation rights, enterprise data architects can now enforce strict least-privilege access across large user populations.

Information security teams gain tighter control over data leakage risks, while BI teams can deploy corporate dashboards broadly without exposing underlying semantic models to unauthorized data extraction or ad-hoc modeling.

Direct Lake flexibility improves schema management and enterprise hybrid models

Data platform managers gain granular control over semantic model updates with the ability to decouple schema synchronization from data refreshes. In addition, Direct Lake models now support table-level conversions to Import mode directly in web modeling, enabling hybrid architectures without breaking model relationships or requiring complex rebuilds.

These updates directly address production stability and cost for large-scale data engineering teams. Decoupling schema changes prevents unexpected source pipeline updates from breaking downstream production reports during routine data loads. Meanwhile, converting specific high-cardinality Direct Lake tables to Import mode optimizes query performance and capacity consumption without sacrificing real-time access to core transactional data.

Modernized local workflows accelerate enterprise BI developer velocity

Power BI Projects (PBIP) now feature instant change reloads without restarting Power BI Desktop, supported by direct integration with Visual Studio Code. Developers can modify report definitions, semantic model metadata, and tracking files in external tools while seeing changes immediately reflected in the local desktop environment.

This update transitions Power BI development from isolated desktop file editing toward mature, continuous integration and deployment (CI/CD) practices. Engineering teams can now enforce standard version control, perform code reviews on PBIP pull requests, and leverage external scripts without experiencing friction from tedious application restarts or broken developer loops.

Business value and architectural impact

The August release moves Power BI closer to a fully governed, enterprise-grade data platform. By balancing strict security access controls with flexible Direct Lake modeling and modern software engineering practices, Microsoft enables organizations to scale their analytics infrastructure efficiently while maintaining control over data integrity and operational risk.

Learn more on Power BI Updates Blog