Browse All Posts (169)
Sandeep S. Nair announces general availability of the DocumentDB extension for Visual Studio Code. The extension combines visual database exploration and an integrated shell, supports Microsoft Entra ID authentication, and provides query analysis, index operations, and local development workflows for Azure DocumentDB and other MongoDB-compatible environments.
Allison summarizes GitHub Copilot updates covering new models, GitHub Enterprise Server access in the Copilot app, and experimental agent workflows in Visual Studio Code. The release notes also cover MCP reauthentication, mobile session handoffs, and a February 2027 deadline for moving from the Local harness to the Copilot harness.
Allison reports that GitHub Enterprise Cloud accounts can now create up to 40,000 budgets, up from 10,000. The higher limit applies automatically on github.com and GHE.com, providing more room for enterprises that assign budgets to individual users or across multiple cost centers and organizations.
Allison outlines the November 16 deprecation of Claude Opus 4.8, Claude Opus 5, and GPT-5.6 Terra across most GitHub Copilot experiences. The announcement identifies replacement models, explains administrator controls for enabling alternatives, and notes exceptions for annual subscribers and runtime refusal fallback.
Microsoft Developer shows how to build and run Azure HorizonDB from Visual Studio Code, keeping setup and AI application development in the same workspace. The video introduces a platform that combines transactions, vectors, AI and orchestration for building intelligent applications.
GitHub explores why NVIDIA’s Auto-Ontology repository gained 8,000 stars in just a few days. The project builds a semantic layer mapping business concepts to database tables, helping AI agents understand data and validate SQL queries, with connections available through MCP, CLI, or APIs.
Allison outlines a public preview update that lets organizations start AI Scans for up to 20 repositories at once. Findings join CodeQL alerts in the code scanning backlog, covering unsupported languages and additional vulnerabilities. Scans can be started from organization settings or alerts pages without a GitHub Copilot license.
Bas Alberts and Xavier René-Corail share lessons from six years of GitHub Security Lab work protecting open source. Their GitHub Universe 2026 session explores software supply-chain security, community vulnerability remediation, and how Dependabot, GitHub Secret Protection, GitHub Code Security, and AI agents help safeguard repositories.
Adam Bogobowicz and Adrian Twarog discuss matching AI models to the task using Microsoft Foundry’s model catalog. They explain why a smaller model can handle the same job for less, making model selection a question of task requirements and cost rather than size alone.
Allison introduces GitHub Mobile support for iPhone Duo, including a layout that adapts when the device unfolds and vertical tabs for navigating repositories and activity. The update is available on iOS, with further refinements planned to make fuller use of the larger screen.
wolfgangdesalvador explains how Azure CycleCloud and Azure Managed Lustre coordinate HPC workloads across regions. The workflow uses Slurm Burst Buffer hooks to stage inputs, run simulations against regional storage, and return outputs, keeping data transfers outside compute allocations while preserving familiar job submissions and application paths.
tamirdresher introduces durable execution for Pydantic AI agents using Azure Durable Task Scheduler. The walkthrough shows how to checkpoint model and tool calls, run locally with the scheduler emulator, and inspect workflow activity in the dashboard, while explaining retry behavior, human approval waits, and limitations around external side effects.
John Savill walks through Azure Enclave, explaining how it starts Azure resources with isolated, secure networking and orchestrates exceptions and required connectivity. The video covers communities, policies, dedicated hubs, approval settings, endpoints, enclave connections, workloads, and pricing, with a focus on network isolation and compliance.
John Savill's Technical Training provides a quick overview of Azure Enclave, a networking solution for Azure described as secure and isolated by default. The video introduces its focus on network isolation and security within Azure.
MinseokSong examines production failures in AI-assisted translation and presents an Azure reference architecture that separates durable requests, bounded execution attempts, candidate artifacts, validation, and human approval. The lessons cover incomplete model output, dependency failures, concurrent work, and progress visibility without confusing successful execution with an acceptable result.
.NET 11 is generally available with two years of support, C# 15 unions and closed hierarchies, and expanded libraries, while .NET 8 and .NET 9 have reached end of support. Azure Functions now supports .NET 11 for C# isolated workers as in-process support ends, making hosting and SDK changes part of the upgrade rather than just a target-framework edit. Aspire 13.6 adds SQLite-backed diagnostic history and a Native AOT dashboard, retaining more telemetry while reducing memory use and startup time in reported benchmarks. Visual Studio 2026 version 18.12 enables SDK-powered Copilot Chat by default, while LTSC 18.11 offers a two-year fixed-feature support window. The releases pair migration requirements with concrete tooling improvements, but Runtime Async, C# memory-safety work, and Blazor AI components remain separate preview capabilities.
Git 2.56.1 fixes six vulnerabilities, with backports and patched GitHub Enterprise Server releases available, but developers must still update bundled Git copies even though GitHub's cloud services are patched. November .NET servicing addresses multiple vulnerability classes, while Windows tightens PowerShell signature verification and PowerShell 7.7 strengthens execution defaults, making compatibility testing part of patch deployment. GitHub added opt-in workflow-log secret scanning and repository-level AI Scan API controls, and announced that Dependabot alerts will pause on most repositories inactive for one year starting December 8 unless policy exemptions apply. Research into poisoned coding-agent models and small weight-level backdoors shows why restored coding performance and ordinary evaluations do not establish model safety. Across agent tooling, Kubernetes policy, and remote-access investigations, the practical message is to verify what controls actually enforce: connectivity
.NET 11 is available as .NET 8 and .NET 9 leave support, making runtime migration an immediate maintenance task, while PowerShell 7.7 introduces script-default and security changes that require compatibility checks. Aspire 13.6 adds persistent dashboard history and lower memory use, alongside broader language hosting and deployment tooling for distributed applications. Git 2.56.1 fixes six vulnerabilities, and GitHub adds opt-in secret scanning for Actions logs while setting a December deadline for ARM32 hosted jobs. Azure Resource Manager MCP now supports resource changes and deployment operations, while Foundry Toolbox guidance separates deployment from validated version promotion. Across these developments, teams need explicit support checks, deterministic test gates, and permission boundaries rather than treating successful automation as proof of production readiness.
GitHub Copilot adds Claude Fable 5.2, Gemini 4 Carbon, and Gemini 4 Hydrogen as six model retirements approach on December 14, making replacement access, usage-based pricing, and data-retention policies immediate decisions for teams. Visual Studio 18.12 makes SDK-powered Copilot Chat the default and brings Bring Your Own Model to general availability, allowing hosted or local models in Agent Mode without GitHub sign-in. Copilot for Xcode gains stable Custom Agents, Subagents, and Plan Mode, while GitHub Enterprise Server support becomes generally available in the Copilot app and enterprise GitHub Apps gain seat-management APIs. Across modernization guidance and agent-assisted engineering examples, the shared emphasis is on explicit permissions, isolated execution, and independent validation rather than treating agent output as proof that a change is safe.
.NET 11 is generally available for C# isolated worker apps in Azure Functions as in-process support ends, requiring project SDK changes and attention to hosting restrictions. The Durable Task extension for Microsoft Agent Framework reaches general availability, preserving execution state outside workers and reusing recorded model results during replay, while Foundry Toolbox guidance separates immutable tool versions from release promotion. Azure Migrate now assesses Container Apps readiness, and Aspire updates add generally available TypeScript apphosts, Kubernetes storage modeling, and dashboard history with development-only security boundaries. Cosmos DB Safe Key Rotation and DocumentDB's generally available service-managed and graceful failovers give teams more control over credentials and regional recovery, though outage failover can lose unreplicated writes. These developments pair broader automation with explicit requirements for validation, authorization, and recovery planning.
New model-security research shows that poisoned post-training data and edits to fewer than ten neurons can implant hidden behavior that ordinary task evaluations miss, making checkpoint audits and backdoor-specific testing important before deployment. Azure opened HDv2 preview registration for data preparation and dense agent sandboxes, while GPU inference tests reported performance in line with reference benchmarks, with preliminary sandbox estimates still requiring workload-specific validation. Windows ML introduced experimental llama.cpp support and task-specific text generation and speech recognition APIs, giving developers another route to local inference without replacing existing ONNX Runtime integrations. OneLake storage support for Unity Catalog reached general availability in Fabric while remaining in preview on Azure Databricks, and Fabric Warehouse previewed Git-based schema deployment alongside an upcoming SQL pool API change that requires client updates. Microsoft's annou
.NET 11 is available with an SDK template for Model Context Protocol (MCP) servers, giving developers a shipped starting point for exposing application capabilities to agents while accompanying Blazor AI components remain experimental. The Durable Task extension for Microsoft Agent Framework reached general availability, allowing workflows to persist execution state, reuse recorded model results during replay, and wait for external approvals without keeping workers active. Azure Content Understanding added GPT-5.6, GPT-6, and GPT-6.1 model choices plus preview agentic extraction, while Microsoft Foundry expanded text and video options with distinct hosting, retention, and access conditions. Azure Speech made multilingual refinement and multichannel transcription generally available, although their compatibility limits mean developers cannot combine all features. Across these releases, security research and reliability evaluations reinforce the need to verify model provenance, constrain
Harald Binkle describes how he uses GitHub Copilot to turn measurements, photos, and video references into printable 3D models. His workflow combines Python for fit-critical geometry, Blender for organic shapes, and iterative visual reviews, while distinguishing digital validation from proof of physical fit.
slindsay brings together 46 community posts, videos, and tutorials covering Microsoft Fabric. The October spotlight includes write-back and forecasting apps, AI-ready semantic models, data agent security, deployment workflows, Spark performance, Direct Lake, Real-Time Intelligence, and Power BI modeling, alongside FabCon recaps and DP-700 study resources.
Cybersecurity researcher Niels Provos discusses the risks of giving autonomous AI agents unrestricted access to email, calendars, and data. This GitHub interview clip examines how missing security boundaries and malicious behavior hidden in third-party agent skills can expose enterprises to vulnerabilities.
Jesse Houwing explains how to protect repositories that publish GitHub Actions using rulesets, immutable releases, reviewed workflows, and locked dependencies. The guide covers maintainer access, token permissions, release verification, and security checks, while explaining where each safeguard falls short and how to keep floating version tags safe.
Jesse Houwing explains how to protect the repositories and release workflows behind GitHub CLI extensions. The guide covers branch and tag rulesets, immutable releases, artifact verification, workflow lock files, and security scanning, while highlighting installation and pinning limitations that maintainers and users need to understand.
Jeff Doolittle of Trimble explains how he used GitHub Copilot to build a repository governance system in four days without manually writing code. The session covers combining custom prompts with GitHub Actions, designing reversible enforcement, and providing self-service remediation while preserving human judgment and reducing developer friction.
April Leonard, Angela Jiang, Richard Seroter, and Rohan Varma discuss where AI-assisted development could head over the next two years. Recorded at GitHub Universe 2026, the panel explores frontier models, autonomous agents, MCP integrations, code security, and the role of human judgment in changing developer workflows.
Carlin Cherry explains how attackers introduce malware into dependencies through compromised maintainer accounts and lookalike package names. The short covers why malicious packages can execute during installation, how the GitHub package firewall blocks known malicious packages, and how Dependabot alerts developers to malware already present in their dependencies.
Jose_Manuel_Jurado explains how to check available Azure SQL Database backups using SQL queries, Azure CLI, and the Azure portal. The guide distinguishes point-in-time restore windows from long-term retained recovery points, explains Hyperscale limitations, and shows why backup metadata alone cannot prove a successful restore or historical policy coverage.
Rafia_Aqil explains how the Fabric OneLake catalog and Microsoft Purview divide data governance responsibilities. The article maps existing governance outcomes to Fabric capabilities, identifies gaps in glossary, scanning, and policy coverage, and outlines a phased transition that validates data access, protection, audit evidence, and AI governance.
Aniket Deshpande, Pouneh Partowkia, and Vinay Maddi explain how BioNeMo skills connect RFD4-Proteina to Microsoft Discovery for natural-language protein binder design. They cover the lightweight tool-container architecture, workspace setup, and the data-residency trade-offs between NVIDIA-hosted inference and deployments inside a customer’s Azure tenant.
Microsoft Security Research, Sagar Patil and Jeremiah Jasmin examine how SCKit abuses ScreenConnect to establish unauthorized remote access and target technician sessions. Their analysis separates server-side extension abuse from client-side propagation and provides remediation guidance, indicators of compromise, and Microsoft Defender XDR hunting queries for investigating affected endpoints.
Mihir22_Patel examines performance benchmarks for AMD-powered Azure confidential VMs, separating generational improvements from the overhead of data-in-use protection. Independent DCas_v6 results and internal DCas_v7 preview tests show gains across workloads, while matched comparisons explain why confidentiality overhead should be assessed against a same-generation baseline.
Mihir22_Patel examines Intel benchmarks and Azure engineering measurements to explain performance trade-offs for confidential VMs with Intel TDX. The analysis separates compute overhead from storage and networking improvements, showing where local NVMe and MANA approach general-purpose VM performance and why workload-specific testing remains necessary.
Allison outlines six GitHub Copilot model deprecations scheduled for December 14, 2026, with suggested replacements and guidance for administrators. The changes affect all Copilot experiences, although Claude Haiku 4.5 remains available to individual annual-plan subscribers. Users should update workflows and integrations before the deadline.
WSilveira walks through migrating NuGet-based Azure Logic Apps Standard applications and custom built-in connectors to .NET 10 without rewriting connector code or workflows. The guide explains project retargeting, dependency fixes, different local and Azure host settings, deployment verification, rollback steps, and a GitHub Copilot skill that assists the migration.
fiveisprime introduces Azure Container Apps sandbox support in the Visual Studio Code extension. Developers can create sandbox groups and sandboxes, execute commands, take named snapshots, and stop or resume resources from the Azure Resources view, with status labels and operation results available inside the editor.
Connor Al-Joundi outlines the transition from Phi Silica to Microsoft Aion Instruct for local AI on Windows, including planned Insider and retail rollouts and expanded GPU support. Developers should test custom prompts, LoRAs, and output parsing against the new model, which introduces a 32K-token context limit and invisible watermarking.