Browse Artificial Intelligence Blogs (53)

John Edward outlines practical ALM and environment strategy guidance for Microsoft Copilot Studio, focusing on how to run copilots like enterprise applications with multi-environment setups, solution-based development, source control, CI/CD pipelines, configuration management, governance, and ongoing monitoring.

Hidde de Smet compares the GitHub Copilot App and the VS Code Agents Window, focusing on how each surface supports agent-first workflows: isolated sessions, worktrees, review/CI loops, and customization via MCP and instruction files. It includes a practical “which one should you use?” decision guide for day-to-day development vs delegated work.

DevClass rounds up Microsoft Build announcements that matter to developers, including new Windows sandboxing for AI agents (MXC), an Arm-based Surface RTX Spark Dev Box, GitHub Enterprise Local for connected or air-gapped environments, Azure Linux updates, and Microsoft-maintained Coreutils for Windows.

John Edward explains how to design multi-agent architectures in Microsoft Copilot Studio, using an orchestrator copilot plus specialized agents (IT, HR, sales, analytics). The article covers communication options (Power Automate, Dataverse, REST APIs), governance and security considerations, and practical scaling guidance like monitoring, shared knowledge sources, and independent versioning.

John Edward walks through creating a first AI agent in Microsoft Copilot Studio, from defining the agent and connecting knowledge sources to enabling generative answers, testing conversations, and publishing to channels like Teams and websites.

John Edward explains where Microsoft Copilot Studio sits in the Power Platform, and how it connects conversational AI to apps, workflows, data, and analytics through tools like Power Apps, Power Automate, Dataverse, and Power BI.

Jesse Houwing shows how to automate GitHub Copilot AI Credits budgeting by assigning per-user budgets based on Microsoft Entra ID group membership, using a GitHub Actions workflow and a PowerShell script that calls the GitHub enterprise billing API via the GitHub CLI.

Hidde de Smet lays out a practical KPI scorecard for teams adopting AI coding agents under usage-based billing, using GitHub Copilot’s AI Credits model as the concrete example. It focuses on measuring speed, quality, reliability, and spend together, with a rollout plan and data sources you can wire into a weekly dashboard.

Randy Pagels explains how to reduce repeated prompting by capturing team conventions in a copilot-instructions.md file so GitHub Copilot can generate code that matches your repo’s standards, architecture expectations, and preferred testing and design patterns.

John Edward explains how Declarative Agents and Autonomous Agents differ in Microsoft Copilot Studio, focusing on how each approach handles control, decision-making, and multi-step work. The article maps the trade-offs (predictability vs flexibility), gives practical use cases for each agent type, and offers guidance on choosing the right model for a given workflow.

John Edward explains how Microsoft Copilot Studio structures conversational agents using topics (intent-focused conversation pathways) and nodes (step-by-step actions inside a topic). The article breaks down common node types, shows how topics and nodes fit together in real flows, and shares practical design tips for building maintainable copilots.

Jesse Houwing breaks down why common “AI ROI” dashboards (tokens, PR counts, lines of code) don’t actually measure value, and how they can backfire through metric gaming and biased attribution. He proposes outcome-based measurement and post-build validation practices that better reflect real impact.

John Edward outlines common enterprise AI agent architecture patterns you can implement with Microsoft Copilot Studio, including single-agent designs, multi-agent orchestration, RAG, human-in-the-loop workflows, and event-driven automation, with notes on integrations, governance, and compliance considerations.

John Edward outlines practical Microsoft Copilot Studio scenarios teams are using to cut repetitive work, including customer support, HR onboarding, IT help desk triage, internal knowledge search, sales lead qualification, and meeting follow-ups across common Microsoft 365-connected workflows.

Harald Binkle demonstrates a practical BMAD workflow using GitHub Copilot to turn fuzzy requirements into reviewable artifacts: a PRD, project context, epics/stories, architecture decisions, risk-based test design, and traceability. The example focuses on enterprise authentication concerns like MFA, tenant isolation, RBAC, and auditability.

Rob Bos argues that as GitHub Copilot shifts to usage-based billing, teams should stop fixating on token costs in isolation and instead measure what they get back: foundation work, reduced tech debt, and faster MVP delivery. He shares real usage patterns, cost concentration among heavy users, and practical steps to manage spend without throttling engineers.

Hidde de Smet breaks down what AI coding agents actually cost once GitHub Copilot switches to usage-based billing, including how credits map to tokens, why model choice changes your bill, and how to budget for agent-heavy teams without surprising finance.

Rob Bos introduces the GitHub Copilot App technical preview and shares a practical first look at using it for repository maintenance, including parallel agent sessions, session modes (Interactive/Plan/Autopilot), and the Agent Merge workflow for handling CI failures, merge conflicts, and security-related alerts.

John Edward explains how GitHub Copilot changes team workflows around pull requests, code review expectations, and knowledge sharing. The article focuses on the trade-offs of faster AI-assisted coding, why review discipline matters more, and how teams can add guardrails like testing and security scanning without losing collaboration.

Jesse Houwing breaks down why GitHub Copilot is moving from Premium Request Units to token-based, usage-based billing, and what that means for model selection, cost predictability, and newer features like Agent Mode, Cloud Coding Agent, and Copilot Code Review—especially for organizations managing budgets and policies.

John Edward outlines an architecture for a “Daily Stand-Up Agent”: a custom AI copilot that pulls sprint activity from Jira and Azure DevOps, detects blockers, and generates consistent stand-up summaries. The post focuses on connectors, grounding ticket data, conversational reporting, and practical considerations like security and data quality.

Rob Bos shares a real-world GitHub Copilot CLI mishap where an unintended Copilot CLI extension caused repeated prompts to close GitHub deployment-status notifications, and explains how he tracked down the source and removed it.

DevClass reports on the Zed editor reaching version 1.0, covering its Rust-based architecture, GPU-accelerated UI, built-in language server support, and the editor’s growing set of AI features (including agents) alongside an option to disable AI entirely.

John Edward explains how Architecture Decision Records (ADRs) capture the “why” behind technical choices, and how AI tools can generate consistent ADR drafts quickly so teams can focus on review, accuracy, and long-term knowledge sharing.

John Edward breaks down the core building blocks of copilot agent systems, explaining how interface, orchestration, LLMs, tools, memory, and safety layers fit together. The article also covers common design patterns like RAG and tool-using agents, plus practical challenges around context, reliability, latency, and security.

Rob Bos shares an overview of his open source projects spanning GitHub and CI/CD tooling, Azure-backed services, security reporting, and local-first AI utilities, with links to each repo and a clear description of what each tool does.

Hidde de Smet shows how to combine five GitHub Copilot customization file types in a single .NET Aspire repo, so the right instructions, skills, prompts, and agent roles load at the right time without bloating every chat request.

John Edward discusses how GitHub Copilot changes programming education, where it can speed up learning, and where it can undermine fundamentals if students rely on it too heavily. The post outlines practical habits for students and classroom approaches for educators to use Copilot without losing academic rigor.

Rob Bos breaks down five GitHub Copilot and agent extensibility surfaces that create supply-chain and governance gaps in large enterprises, and explains what controls exist today (and where they don’t) across Copilot CLI plugins, APM, gh skill, MCP servers, and VS Code extension registries.

Hidde de Smet's Blog breaks down the difference between AGENTS.md (repo-wide, always-on instructions for coding agents) and .agent.md (custom agent profiles for GitHub Copilot), including where to place each file, what fields matter, and how to use roles, tool restrictions, and handoffs safely.

John Edward explains why event-driven architecture is a strong fit for agentic AI systems, and breaks down the core patterns (pub/sub, event sourcing, sagas) plus practical concerns like ordering, observability, and infrastructure overhead.

Hidde de Smet's Blog explains how GitHub Copilot “skills” work via SKILL.md folders, why the YAML description is the key to discovery, and how this approach keeps context lightweight compared to a giant copilot-instructions.md. It includes a practical Azure Monitor/Application Insights KQL skill you can copy into a repo.

DevClass.com reports on Visual Studio 18.5 (Visual Studio 2026), covering new Copilot-driven “agentic” debugging, changes to how IntelliSense/Copilot suggestions are prioritized, and ongoing developer complaints about theme contrast and forced auto-updates.

Hidde de Smet compares three AI coding setups—single-agent, agent-with-tools, and multi-agent—using a realistic .NET Aspire + ASP.NET Core rate-limiting task to show trade-offs in fit, cost, latency, and common failure modes.

John Edward explains when to use single-agent vs multi-agent AI architectures in a Microsoft context, mapping common designs to Semantic Kernel, AutoGen, and Azure services like Azure OpenAI, Azure AI Search, Functions, Service Bus, and AKS.

DevClass.com reports on GitHub’s private preview of Stacked PRs, a workflow for breaking large changes into smaller, independently reviewable pull requests that can still depend on each other, with an optional gh stack CLI that’s also intended to work well with AI agents.

Jesse Houwing summarizes GitHub’s update that GitHub Copilot can now keep inference processing and associated data within US or EU data residency regions, and shows the enterprise/org policy you must enable to restrict Copilot to data-resident models.

Rob Bos walks through running GitHub Copilot CLI against local OpenAI-compatible inference servers (Ollama, LM Studio, Foundry Local, vLLM/TGI), focusing on the practical constraints (32k context, tool calling, VRAM/KV-cache) and sharing concrete Windows/PowerShell setup and throughput numbers.

Emanuele Bartolesi shows how to point GitHub Copilot CLI at an Azure AI Foundry (Azure OpenAI) deployment using a BYOK-style setup, including how to deploy a model, build the correct endpoint URL, set the required environment variables, and validate the connection.

Emanuele Bartolesi explains how to run GitHub Copilot CLI against a local LLM via LM Studio’s OpenAI-compatible API, including the exact PowerShell environment variables needed to avoid cloud fallback and when this offline setup is (and isn’t) worth using.