Browse All Posts (132)

Sindhura Bandhakavi explains why “using fewer tokens” can backfire when you’re building AI agents: if the agent has too little context, it may compensate by making more calls to reach the same accuracy, which can increase total cost.

Bruno Capuano and Luis Quintanilla explain how .NET’s AI building blocks fit together for real-world apps, focusing on data ingestion, embeddings, and vector search to ground LLM responses in your own documents and business data.

sbaynes reports on Microsoft Research work proving the KLS conjecture, a long-standing result in high-dimensional geometry with implications for sampling, optimization, and randomized algorithms. The post also describes a human-directed, agentic AI workflow used to explore candidate proof paths and support derivations and verification.

Microsoft Security Research and Microsoft Threat Intelligence detail how Storm-0940 used the ARTEX agentic AI framework to accelerate a ConnectWise supply chain compromise, then abused OAuth application identities to access Microsoft 365 tenants via app-only authentication and Microsoft Graph, with concrete mitigation steps, detections, IOCs, and hunting queries.

prl explains the preview incremental migration workflow in Migration Assistant for Microsoft Fabric Warehouse, focusing on migrating database object definitions in phases, re-running migrations without recreating the destination, and handling naming conflicts safely. The post also clarifies how object migration differs from data loading and where Data Factory copy jobs fit in.

GitHub shares a September platform update roundup covering Project HydraFusion’s expansion to VS Code and the GitHub Copilot app, plus new workflow features like local sandboxing GA, automated PR approvals that can satisfy merge gates, and bulk autofix for CodeQL findings.

Niels Laute announces PowerToys 0.102 and walks through the headline changes, including new mouse utilities, a quick-access shelf for Command Palette, ZoomIt snip dictation, and new command-line tools that make it easier to script and automate PowerToys features from the terminal.

prl introduces the preview of Fabric Migration Assistant for migrating Teradata workloads to Fabric Data Warehouse, focusing on translating metadata and SQL/BTEQ into Fabric-compatible T-SQL, reviewing exceptions, and choosing practical data-movement and validation approaches to reduce manual migration effort.

QixiaoWang introduces the Fabric Notebook Toolkit (FNTK) preview, a CLI that lets coding agents interact with live Microsoft Fabric notebooks—reading and editing cells, running code, and pulling outputs and Spark diagnostics—so iterative debugging and validation can happen directly from an agent-driven terminal workflow.

Allison announces a public preview in GitHub CLI v2.103.0 that lets you opt in to expiring, auto-refreshing OAuth credentials, reducing the risk of long-lived leaked tokens. The post explains how to enable the feature, how it behaves with Git over HTTPS, and what to watch for in scripts that cache tokens.

Eleanor White explains how GitHub is moving its Git storage to Azure, focusing on the last ultra-busy repositories blocked by cross-region latency. The post breaks down per-reference write quorums and a “virtual replica” design that makes Azure Blob Storage part of the durability path so reads and writes can scale independently.

Laura Jiang announces Agentic Autofix for GitHub Advanced Security in Azure DevOps, which generates proposed fixes for both code scanning alerts and vulnerable dependencies. The update uses repository context plus custom instructions/skills to better match team standards, and adds triage signals like EPSS and dependency scope to help prioritize what to fix first.

John Edward breaks down six Copilot Studio enhancements landing in general availability on October 26, 2026, focused on building and operating Copilot Chat (declarative) agents with more predictable conversation flows, better testing and evaluation, usage monitoring, and clearer lifecycle controls across environments.

Leah Tran introduces a new Visual Studio 2026 (18.11) Git workflow feature that lets you compare a historical commit directly against your current working tree, without checking out the old commit or switching branches, making it easier to inspect diffs that include uncommitted changes.

Visual Studio Code highlights improvements to the built-in Markdown editor, focusing on a smoother writing and editing workflow inside VS Code.

Ben Sickle outlines Microsoft’s announcements at NVIDIA GTC Berlin 2026 focused on running AI across Azure and customer-controlled environments, including Azure Local sovereign deployments, Foundry Local for on-prem model inference, confidential VMs, open-model endpoints for NVIDIA Nemotron via Fireworks AI on Azure, and GPU-accelerated analytics in Microsoft Fabric.

The Visual Studio Code Team shares the 1.144 Insiders changelog, including updates to the experimental session composer, richer Mermaid diagram controls in the Markdown editor, improved OAuth token isolation for MCP servers, and an early feature for comparing multiple models without opening a folder.

DivSwa explains how to move off Microsoft Foundry workflows (retiring Dec 1, 2026) by orchestrating Foundry Prompt Agents and Hosted Agents inside Azure Logic Apps, combining event-driven triggers and connectors with deterministic workflow steps around agent calls.

Johnson Shi explains how AKS GPU memory profiling helps you move from “GPU memory pressure” alerts to the exact functions responsible for allocations, using eBPF-based tracing and Grafana flame graphs so you can right-size GPU memory and validate fixes with repeatable tests.

Anna Hoffman explains how Microsoft SQL spans on-premises, Azure, and Microsoft Fabric, highlighting what’s new in the engine, developer tooling, and migration options. She also demonstrates AI-assisted performance troubleshooting and scalable vector search, and shows how Azure Arc/Azure Local and Foundry-based experiences can support modern data-driven apps.

Sukhwant Kaur demonstrates how SQL database in Microsoft Fabric can support app development, analytics, and AI in one place—covering database creation, workload migration, and automatic mirroring into OneLake, plus Git-based workflows with VS Code and a T-SQL RAG example for product search.

David Pless explains how Azure SQL Database Hyperscale’s cloud-native architecture lets teams scale compute, storage, and read capacity independently, and how features like serverless compute, elastic pools, named replicas, and snapshot-based backups map to different workload and availability needs.

Patrick LeBlanc demonstrates Fabric Database Hub’s database agent for observing SQL Server databases, surfacing performance signals, and guiding investigation from detection through hands-on troubleshooting, including a blocking scenario in SSMS and follow-up analysis workflows in VS Code with GitHub Copilot support.

Borko Novakovic explains how Azure SQL Managed Instance (SQL MI) aims to make it easier to move SQL Server workloads to Azure while modernizing over time, with demos around replication and minimal-downtime migration plus a look at new performance configuration options and update-policy trade-offs.

Pam Lahoud explains how to migrate SQL Server workloads to Azure Virtual Machines while keeping OS and database control, with practical guidance on VM sizing, storage selection, and cost considerations like constrained cores for licensing. She also outlines what to build next using Azure services and related Microsoft data/AI capabilities.

Bob Ward explains what “AI-ready” means in SQL Server 2025, focusing on native vector capabilities, embedding generation, and RAG patterns you can implement directly with T-SQL across local and cloud AI model deployments.

Amit Khandelwal demonstrates an end-to-end migration of a working app from SQL Server to Azure SQL Database Hyperscale, using SSMS assessments and VS Code tooling to evaluate readiness, address blockers, and execute a BACPAC-based move. He also shows GitHub Copilot helping resolve a Service Broker issue and validating the cutover.

Alicia Li summarizes Q3 2026 updates to Microsoft Fabric Eventstreams, focusing on new managed and custom connectors, CDC ingestion from operational databases, schema registry and schema-aware processing for mixed event types, in-flight enrichment with Delta tables, improved monitoring and diagnostics, and stronger deployment and networking controls.

naafiahsadique introduces Azure Copilot for Azure Local, an AI-powered, agentic troubleshooting experience aimed at helping operators investigate and resolve issues across distributed Azure Local infrastructure using natural-language questions and guided, multi-step diagnostics.

Visual Studio Code introduces Canvases (Preview), a shared workspace concept in VS Code designed to help teams plan releases, prioritize backlog work, and collaborate with a coding agent using contextual cards that can be turned into actionable plans.

jeburke announces general availability of two new Azure Container Registry platform metrics—Data Transfer and Transactions—and explains how to use Azure Monitor and the integrated Grafana dashboard to break down registry traffic by API, region, and response type to investigate throttling, failures, and outbound bandwidth changes.

Allison announces a GitHub repository governance update that lets you convert all classic branch protection rules into repository rulesets in one action, making it easier to standardize and manage branch policies at scale across branches and organizations.

Allison summarizes recent improvements and bug fixes in GitHub secret scanning, including an upcoming audit log field change that may affect integrations, plus updates to token revocation support, webhook payloads, and several API and push-protection edge cases.

tushar-msft summarizes January–August 2026 Graph Query Language (GQL) updates in Microsoft Fabric Graph, covering new path-finding and traversal controls, path inspection functions, multi-stage query composition, richer aggregation options, and added value/text functions (casting, ranges, Unicode normalization, and regex helpers).

Allison announces an update to the Dependabot alerts REST API that adds a new version_requirement field, making it easier to retrieve the dependency version (or constraint) recorded for an alert without parsing lockfiles or issuing separate GraphQL queries.

jovanpop-msft announces four preview T-SQL enhancements in Microsoft Fabric Data Warehouse—GROUP BY ALL, ORDER BY ALL, QUALIFY, and the FROM ... SELECT query format—aimed at reducing repetitive SQL, simplifying maintenance when projections change, and making analytical query logic easier to read.

alyssaschimm announces the GA of advanced platform metrics in Azure Monitor, starting with Azure Storage. The update adds container-level metrics (capacity and blob count), shows where to enable and query them in the Azure portal and Metrics Explorer, and outlines pricing plus upcoming expansion to more clouds and resource providers.

Allison announces GitHub package firewall in public preview for GitHub Enterprise Cloud, a supply-chain security feature that inspects package metadata and downloads to block malicious dependencies before they reach CI/CD hosts or developer workstations.

Julien Dubois shares practical stories from three Java Champions using GitHub Copilot in real projects, focusing on agentic workflows, verification gates, and staying accountable for what ships—from large content migrations to debugging tricky concurrency issues on Raspberry Pi.

jovanpop-msft announces a preview update for Microsoft Fabric Data Warehouse and the SQL analytics endpoint: new and enhanced T-SQL analytical functions for distribution analysis, including exact and approximate median/quantile calculations and expanded percentile aggregate syntax.