Azure Weekly

Issue #549 - 1st February 2026

Highlights this week include: Azure Static Web Apps Skill for GitHub Copilot – A new GitHub Copilot skill that reduces Azure Static Web Apps deployment time from 30+ minutes to under 3 minutes through guided framework detection, configuration, and deployment. Azure Databricks Agent Bricks Knowledge Assistant (GA) – Now generally available, this feature empowers you to create, deploy, and manage AI agents directly within the Azure Databricks platform, using unified data and AI capabilities.

Secure AI Agents with Microsoft Entra Agent ID – Microsoft's new Agent ID (public preview) provides identity, governance, conditional access, and lifecycle management for AI agents through a unified registry across Copilot Studio, Azure AI Foundry, and Security Copilot. Microsoft Purview Data Security Investigations (GA) – Now generally available, this tool uses AI-powered search, semantic embeddings, and risk scoring to investigate data security incidents at scale with new purge mitigation capabilities. Azure Savings Plans Explained – A clear breakdown of Azure Savings Plans offering savings through hourly commitments, covering the use-it-or-lose-it model, discount calculations, and practical guidance on when they make sense.

Fabric January 2026 Feature Summary – Highlights include Item Reference variables for unified configuration, direct VS Code notebook editing, AI-powered catalog experiences, and enhancements across Data Engineering, Warehouse, and Real-Time Intelligence. Fabric Connection inside Notebook (Preview) – Notebooks can now use cloud connections (Azure Blob Storage, PostgreSQL, S3, Key Vault, and more) directly via Get Data, with code snippet generation for accessing external data sources. Workspace Outbound Access Protection for Data Factory and OneLake Shortcuts (Preview) – Extends workspace-level outbound access protection to Pipelines, Copy jobs, Dataflows, and OneLake Shortcuts, enabling granular control over external connections.

If you're interested in all things Microsoft Fabric - don't forget to sign up for our new newsletter - Fabric Weekly - which we'll start publishing in the next month or so. We'll be moving all Fabric content over from Azure Weekly to Fabric Weekly, just as we did with Power BI Weekly 7 years ago.

Finally, Barry Smart has written a 4-part series which covers how we (endjin) have improved our data engineering approach in the past year by migrating workloads from PySpark / Pandas to Polars: Polars: Faster Pipelines, Simpler Infrastructure, Happier Engineers – Makes the case for Polars as the default DataFrame choice, delivering 5-20x faster performance than Pandas with 8x less energy consumption for most analytical workloads. Under the Hood: What Makes Polars So Scalable and Fast? – Deep dive into Polars' Rust foundation, Apache Arrow columnar storage, lazy evaluation with query optimization, and vectorized SIMD execution. Practical Polars: Code Examples for Everyday Data Tasks – Hands-on examples covering data loading, expressions, filtering, lazy evaluation, DuckDB integration, and streaming execution for larger-than-RAM datasets. Polars Workloads on Microsoft Fabric – Practical guide to running Polars in Fabric notebooks, including path construction, storage options, Delta Lake integration, and tips for code portability between local and cloud environments.