What’s new in Microsoft 365 Copilot for U.S. Government Customers
August 4, 2026
How Microsoft's DSC engine is shaping the new adapter model
August 4, 2026
In this issue:
Ace Aviator of the Month
August 2026’s Ace Aviator: Sonny Gillissen
LinkedIn: https://www.linkedin.com/in/sonnygillissen/
What’s your role and title? What are your responsibilities?
Integration Architect / Cluster Lead
I’m responsible for running Integration projects at our customers, and next to that I’m leading our team of Integration Experts by embedding our mission and vision deeply in the roots of our team.
Can you give us some insights into your day-to-day activities and what a typical day in your role looks like?
My day typically consists out of working together with my customer to validate our integration solutions and form next steps to provide the best fit. This could be meeting with the business to gather requirements, or check in with the team to find the best approach. From an internal perspective I’m working on our go-to-market to keep it aligned with the company’s vision and movements in the market, together with our team of Integration Experts.
What motivates and inspires you to be an active member of the Aviators/Microsoft community?
The fact that it such an active community, and people are very much willing to help each other is what drives me te be an active member too. Especially when you could help someone with your expertise is what really makes my day and provides all the energy to keep outperforming myself, every single day.
Looking back, what advice do you wish you had been given earlier that you’d now share with those looking to get into STEM/technology?
You don’t have to do everything alone. When working with techhology, it may feel like you’re the only one at times, especially when you’re hitting that one niche problem. But in fact, others can have such a positive impact, so keep asking.
What has helped you grow professionally?
First of, I really believe having a team of experts around you that you can collaborate with helped a lot. But on the other hand, actively sharing knowledge and digging into problems of others is what really helped me grow. Not only professionally, also as person.
If you had a magic wand that could create a feature in Logic Apps, what would it be and why?
This question was easier to answer when there wasn’t a thing like Logic Apps Automation, as I believe this was my magic wand before: putting the power of integration in everyone’s hand. But, with that said, I think it would be cool of you can have an “Logic Apps Agent” that creates and maintains your envisioned integration by itself, or with it’s Logic App co-workers (of course with some checkin’ in to you at times). In my opinion this would really bring the power of Logic Apps to literally everyone.
News from our product group
AI Gateway tier of API Management now in public preview
The public-preview AI Gateway tier of Azure API Management provides a purpose-built experience for publishing and governing AI models, MCP servers, and tools. It supports models from Microsoft Foundry and other providers, connector-backed tools, card-based governance policies, Azure RBAC, and OpenTelemetry token-usage metrics.
Changing the engine while the plane is flying: migrating 60,000 apps under live load
This engineering account describes migrating roughly 60,000 Integration Account Function apps from end-of-life runtimes to Azure Functions v4 isolated worker. The team used shadow traffic, parity checks, progressive regional rollout, configuration-based rollback, privacy-preserving telemetry, and a stop-before-delete retirement process.
Hybrid Logic Apps on RKE2: a self-managed cluster with MetalLB
This walkthrough shows how to deploy Azure Logic Apps Hybrid on a self-managed RKE2 Kubernetes cluster using MetalLB. It covers Azure Arc, the Container Apps extension, custom locations, connected environments, ingress, and RKE2-specific fixes involving inotify limits, CoreDNS configuration, and the missing kube-dns service alias.
Announcing Flat File Schema Generation support in Azure Logic Apps Standard
Azure Logic Apps Standard now provides a preview built-in action that generates BizTalk-compatible flat-file XSD schemas at runtime from sample payloads. It supports common delimited and fixed-width formats and can feed generated schemas into Flat File Encoding or Decoding actions.
Hybrid Logic Apps Deployment on Red Hat OpenShift
This guide explains how to deploy Azure Logic Apps Hybrid on self-managed Red Hat OpenShift or Azure Red Hat OpenShift. It covers Azure Arc, SMB storage, OpenShift security context constraints, the Container Apps extension, custom locations, connected environments, ingress choices, DNS configuration, and troubleshooting.
Coding with Logic Apps Standard: Local Functions
This article introduces Local Functions in Azure Logic Apps Standard, which let developers write, debug, and deploy custom .NET code alongside workflows in the same project. Workflow-scoped code can share the Logic App’s deployment, scaling, security, and operational boundary without requiring a separate Azure Functions resource. Common scenarios include message validation, payload enrichment, custom parsing, business rules, and BizTalk modernization. The article also explains when Local Functions are preferable to independently hosted Azure Functions and how they fit into a unified CI/CD process.
News from our community
What’s New with Logic Apps
Post by Gabriel Yang
The article reviews how announcements from Integrate 2026 move Azure Logic Apps toward a central role in enterprise automation and AI orchestration. It covers Logic Apps Automation, AI-assisted workflow authoring, Knowledge as a Service, direct Azure AI Foundry Agent invocation, the Logic Apps Standard SDK for C#, and Azure Connector Namespace access from custom applications. Together, these capabilities reduce infrastructure and integration complexity while retaining governance and scalability. The broader value is a more accessible orchestration layer connecting enterprise systems, knowledge, agents, and operational business processes.
Monitoring Is Not Reconciliation
Post by Al Ghoniem, MBA
The article distinguishes technical monitoring from business reconciliation in enterprise integrations. Successful Logic Apps runs, Service Bus metrics, and API responses show that known work executed, but cannot prove every expected transaction reached the target correctly. Effective reconciliation compares expected and actual states using independent reference sets, stable business identifiers, appropriate timing windows, and risk-based matching depth. It also requires liveness checks, explicit ownership, and controlled recovery decisions. The approach helps detect missing, duplicate, late, or inconsistent transactions before customers, auditors, or downstream controls expose them.
Power Automate’s Big Brother – Azure Functions Connector Namespaces
Video by Sean Astrakhan
The video demonstrates using Azure Functions connector namespaces to trigger Python code from services such as Dataverse, Outlook, and SharePoint. A Dataverse record-creation scenario invokes a function that populates an expiration date, with GitHub Copilot helping author, deploy, and test the solution. The example also highlights the need to direct Copilot toward connector namespaces instead of older HTTP-trigger and webhook patterns. This approach combines managed connector access with pro-code flexibility, reducing separate flow and connection requirements while making Azure Functions more approachable for integration developers.
Building Stateful Agentic AI Workflows with Azure Standard Logic Apps
Post by Sakshi Mittal
The guide explains how stateful agentic workflows extend Azure Logic Apps Standard beyond fixed API orchestration. It compares autonomous agents for repetitive background tasks with conversational agents for interactive scenarios, then outlines prerequisites such as AI model access, connectors, RBAC, managed identities, and Application Insights. It contrasts adaptive agent state with conventional workflow run state and reviews benefits including automation, scale, and contextual decisions. It also highlights trade-offs around cost, nondeterminism, debugging, privacy, latency, and governance, helping teams choose an appropriate pattern before production.
Logic Apps Automation Preview
Post by Steef-Jan Wiggers
This assessment examines Logic Apps Automation as a managed, single-tenant experience with Microsoft-managed capacity, a project-and-application hierarchy, agent tooling, sandboxed code execution, and real-time run history. A failure-triage example shows how agents can use runbook knowledge and structured outputs while deterministic branches control tickets and retries. The article also identifies preview limitations, including missing CI/CD, ownership transfer, audit visibility, and uncertain network support. Its central recommendation is to pilot nonregulated workloads while defining project ownership, app-level access, governance, and deployment requirements before broader enterprise adoption.
Azure Logic Apps Standard | Testing Series
Post by Andrew Wilson
The article introduces a practical testing series for Azure Logic Apps Standard, addressing reliance on manual runs and run-history inspection as workflow estates grow. It proposes layered validation covering testable workflow design, mocked action outputs, unit tests derived from workflow definitions in Visual Studio Code, integration tests, and agent-assisted testing. The approach emphasizes deterministic boundaries, separation of orchestration from connector-heavy implementation, observability through tracking and correlation, and behavior-focused environment parity. These practices aim to create faster feedback, clearer regression detection, and repeatable confidence across frequent workflow changes and deployments.
Mastering Error Handling and Retry Design in Logic Apps
Post by Parth T.
Distributed workflows routinely face timeouts, throttling, network interruptions, and unavailable dependencies, making resilience a core Logic Apps design concern. The article outlines how to identify likely failure points, use fallback paths and compensating actions, and apply bounded retries with exponential backoff and idempotent operations. It also emphasizes thorough logging, correlation identifiers, stakeholder notifications, and deliberate testing of failure scenarios such as invalid responses and partial execution. These practices help teams prevent cascading failures, improve diagnosis, and maintain stable business processes when transient or permanent faults occur.
Introduction to Knowledge Base as a Service KBaaS | Built-In Knowledge for Azure Logic Apps
Video by Srikanth Gunnala
Building retrieval-augmented agents normally requires document chunking, embeddings, a vector store, retrieval logic, and orchestration. This video introduces Knowledge Base as a Service in Azure Logic Apps, which manages those components behind a built-in knowledge-base experience. It demonstrates connecting Azure Cosmos DB and Azure OpenAI, uploading documentation, and attaching the resulting knowledge base to a conversational agent. An internal API documentation assistant provides the practical scenario, showing how an agent can answer questions with responses grounded in uploaded source material while reducing custom RAG infrastructure and setup work.
Knowledge Base as a Service in Azure Logic Apps
Post by Steef-Jan Wiggers
Knowledge retrieval for Logic Apps agents has traditionally required a separately configured search index, indexer, data source, chunking strategy, and embeddings pipeline. This walkthrough explains how the preview Knowledge Base as a Service capability instead parses, chunks, summarizes, vectorizes, and stores uploaded content through Azure Cosmos DB and Azure OpenAI. An HR policy agent demonstrates grounded answers, citations, and appropriate fallback for out-of-scope questions. The article also provides deployment steps, a Bicep-based sample, and practical notes on portal persistence, connection configuration, model availability, authentication, and workflow schema requirements.
Vibe Coding Logic Apps in Azure: Automating Complex Workflows with AI Assistance
Post by Marcel Broschk
The article examines how natural-language, AI-assisted development can accelerate Azure Logic Apps workflow creation without replacing engineering judgment. It covers the native workflow assistant, GitHub Copilot scaffolding, structured prompts, agent-and-workflow patterns, MCP servers, and AI-oriented designs such as retrieval-augmented generation. It also recommends repository-level Copilot instructions, automated tests, static analysis, security controls, and human review. The approach helps integration teams move from intent to deployable workflows faster while retaining responsibility for architecture, authentication, error handling, governance, and operational reliability.
Are Azure Logic Apps really low-code or no-code?
Post by Chris Bradshaw
The article evaluates whether Azure Logic Apps is genuinely no-code and concludes that low-code is the more accurate description. Simple connector-based workflows may require no coding, but production integrations commonly involve expressions, JSON editing, error handling, transformations, infrastructure as code, deployment pipelines, and governance. It offers practical guidance for choosing Logic Apps for orchestration and visible workflows, Azure Functions for custom logic and performance, or a hybrid architecture combining both. This framing helps teams match each service to the problem rather than relying on marketing labels.
Azure Logic Apps at Integrate 2026: The Announcements
Post by Steef-Jan Wiggers
The article reviews five Azure Logic Apps announcements from Integrate 2026 and their implications for integration architects. It covers Logic Apps Automation, Knowledge as a Service, native Azure AI Foundry agent integration, the Logic Apps Standard SDK, and Azure Connector Namespace. Together, these capabilities simplify managed workflow hosting, retrieval-augmented generation, agent orchestration, code-first C# development, and connector reuse outside the workflow runtime. The analysis positions Logic Apps as an enterprise AI connectivity and orchestration layer and recommends revisiting hosting, knowledge retrieval, and SDK adoption decisions.
Azure Logic Apps Agent Loop Production Operations
Post by Steef-Jan Wiggers
Production agent loops require more than successful workflow runs. This article explains how Standard Logic Apps can use Application Insights, run history, and KQL queries to monitor requests, dependencies, exceptions, tool calls, token use, and execution duration. It compares Standard and Consumption pricing, outlines tool, throttling, and conversation-history limits, and describes repeatable deployment through source-controlled workflow definitions, zip deployment, Azure CLI, and environment-specific settings. The guidance helps teams evaluate operational readiness, cost behavior, observability, and deployment practices before moving agentic workflows into production.
Event-Driven Automation
Post by Uttam Chaturvedi
File-arrival notifications can be automated with an Azure Logic Apps Consumption workflow. This walkthrough creates a private Blob Storage container, configures a blob-created trigger, sends file metadata to a REST endpoint through an HTTP action, and emails the team through Office 365 Outlook. It also covers resource organization, testing, run history, security, pricing, and possible extensions such as Teams notifications, SQL persistence, Azure Functions, approvals, and CI/CD deployment. The pattern demonstrates event-driven integration without manually operated scripts or continuously running servers.
Azure Logic Apps Tracking Properties – A Complete Team Guide (Part 2)
Post by Sandro Pereira
Tracking Properties improve Logic Apps observability by adding relevant business and technical context to action telemetry. This guide shows how to configure them on individual actions through the designer’s Settings pane or the trackedProperties section in code view. Values may be static, dynamic, or combined, although expressions must be entered manually without IntelliSense and should be validated carefully. Recommended practices include tracking useful identifiers, standardizing property names, keeping values concise, and excluding credentials or personal data, making Log Analytics queries, dashboards, monitoring, and troubleshooting more consistent.
Navigating Cost Pitfalls in Logic App Consumption Plans
Post by Parth T.
Consumption-based Logic Apps can accumulate unexpected charges through frequent polling, premium connectors, redundant actions, large loops, unnecessary runs, retries, API calls, and oversized payloads. This article recommends event-based triggers, appropriate polling intervals, trigger conditions, simpler workflow logic, filtered or batched processing, and reduced connector calls. It also advises continuous monitoring with Azure Monitor and Cost Management, budgets and alerts, periodic execution reviews, managed identities, and documented architectures. These practices support more predictable spending while preserving workflow performance, scalability, governance, and operational visibility.
Streamlining Logic Apps: The Importance of Versioning and CI/CD
Post by Parth T.
Reliable Logic Apps delivery requires disciplined change tracking and automated deployment rather than manual updates. This article explains how version control supports traceability, collaboration, audits, troubleshooting, and rollback, while CI/CD validates, tests, and promotes workflow changes consistently across environments. Recommended practices include semantic versioning, Git repositories, feature branches, ARM or Bicep infrastructure definitions, parameterized environment settings, Key Vault secrets, approval gates, release notes, monitoring, and rollback plans. Together, these methods reduce deployment risk and downtime while improving release speed, consistency, governance, and maintainability.