Compiled by the Microsoft Fabric Blog team, this detailed feature summary explores September 2025 advancements, including new AI, ML, security, and developer capabilities to empower data professionals.

Microsoft Fabric September 2025 Feature Summary

Overview

This release provides a sweeping upgrade to the Microsoft Fabric platform, including data engineering, governance, AI/ML, developer tooling, and security best practices. Below are the key new features and enhancements covered.

Certification, Events, and Community

  • Special discounted certification exams for Analytics and Data Engineers at FabCon Vienna.
  • Power BI DataViz World Championships announced.

Fabric Platform: Engineering, Management & Governance

  • Parent-Child Hierarchy in OneLake Catalog: More intuitive navigation and understanding of relationships among Lakehouses, Warehouses, and Eventhouses.
  • Govern Tab & Domains Public APIs: General availability of data governance tools, including API-driven management of domains for robust access controls and compliance.
  • Microsoft Purview Integration: Automated sensitivity labels, Data Loss Prevention (DLP) policies, and granular domain-level data protection.
  • Deployment Pipeline Upgrades: New design for deployment management, with improved variable library integration.
  • Terraform Provider Extensions: Expanded resource coverage for infrastructure-as-code, supporting standardized and automated deployments.
  • Fabric CLI Open Sourced: The Fabric CLI has opened to community development, supporting automation and AI-driven data estate management.
  • Fabric Extensibility Toolkit (Preview): Accelerates creation of custom data apps and integrations.
  • Workspace-level Workload Assignment: Granular control over workloads at the workspace level.

Data Engineering

  • User Data Functions (Python): Now GA with features for business logic processing, pandas/async support, OpenAPI generation, and native integration with Notebooks and Pipelines.
  • Python Notebook Enhancements: Multiple kernels, advanced Intellisense with Pylance, lakehouse integration, pandas/Polars/Scikit-learn support, semantic links, resource monitoring, multi-source Git/version checkpoints.
  • Materialized Lake Views Improvements: Optimized refresh, lineage tracking, custom environments, on-demand execution.
  • NotebookUtils API: CRUD programmatic management and high-performance file operations.
  • Spark Monitoring API and Analysis Tools: Deep resource metrics, performance tuning, outlier detection for Spark jobs.

Data Science & ML

  • Data Agent Integrations: NL2SQL, mirrored databases, and Python SDK provided.
  • LLM-Driven Feedback: New evaluation tools for example queries and transparency features.
  • Data Wrangler AI: Apply AI-driven enrichment, code generation, and translation with Copilot; PROSE-based smart suggestions enable low/no-code transformations.

Data Warehousing/Databases

  • MERGE Transact-SQL: Complex table operations streamlined for ETL jobs.
  • Migration Assistant: AI-driven migration assistant for moving from Synapse SQL Pool/SQL Server to Fabric Data Warehouse.
  • MSSQL Extension for VS Code: Full-provisioning and in-editor development (Preview).
  • Point-in-Time Restore Upgrades: Recovery window extended to 35 days for more robust operational recovery.
  • Git Integration: System object references and shared query tracking enabled for collaborative development.
  • Performance Dashboard: Now tracks memory usage in addition to existing metrics.

Real-Time Intelligence

  • Maps in Fabric: Integrated geospatial visualization tools.
  • Azure Monitor Logs: Native, real-time streaming of Azure diagnostics into Fabric analytics via Eventstream.
  • Eventstream Security: Workspace Private Link ensures secure, private data ingestion; anomaly detection and scaled-up Activator for 10,000 events/sec.

Data Integration & Orchestration (Dataflow Gen2, Pipelines, Mirroring, Gateways)

  • Pipelines: Simplified terminology, enhanced scheduling, and improved Dataflow/Pipeline integration.
  • Dataflow Gen2: GA for incremental refresh to Lakehouse; CI/CD/param support; Copilot enhancements for code generation and explanation; improved partitioning and modern evaluation engine; extended support for public parameters and new connectors.
  • Pipelines Activities: Improved integration for scheduling, debugging, and invoking pipelines; full support for On-premises/VNet gateways for secure orchestration.
  • Copy job and Orchestration: Support for CDC, incremental copy, parameterization, and new connectors; automation with the variable library.
  • Mirroring: Continuous replication from Google BigQuery, Oracle, Azure SQL MI/DB with firewall and workspace identity support.
  • Gateways: Secure data movement with VNET Data Gateway for pipeline/copy job.

Developer Tooling and Extensibility

  • VS Code Extension (GA): Multi-workspace management, Git source control, integrated SQL access.
  • Open Source Integrations: Fabric CLI, Extensibility Toolkit, and robust API surface enable script-driven development and deployment.

Security & Compliance

  • Microsoft Purview: DLP, default sensitivity labels, and automated security enforcement.
  • OneLake Secure Tab: Unified view of security roles and user access.
  • Workspace Private Link: VNET integration secures connections between data sources and Fabric.

More

  • Enhanced connection and role management in connectors for SAP, Salesforce, Snowflake, DB2, and others.
  • Support for advanced schema mapping and delta tables.
  • General usability improvements, debugging support, new pricing for Dataflow Gen2, and expanded REST APIs

References

For specific details, usage guidance, or new feature walkthroughs, refer to the individual sections and follow related documentation links.

This post appeared first on “Microsoft Fabric Blog”. Read the entire article here