Tag: ai-agents
32 posts tagged with "ai-agents"
-
Foundry Routines Ends the Agent Glue-Code Tax. That Was the Whole Point.
November 4, 2026
Foundry Routines brings scheduled and event-driven execution into Azure AI Agent Service, collapsing the Logic App and Function stack teams bolt onto every production agent.
-
Copilot Studio Runtimes: One Studio, Two Execution Models, and the Wrong Pick Costs You in Production
November 2, 2026
Copilot Studio runtimes: how agents and agent flows execute and fail differently, plus the rule: if the happy path fits numbered steps, build a flow.
-
The Microsoft Agent Framework Harness Is GA: What It Ships, and What You Still Prove
October 30, 2026
The Microsoft Agent Framework Harness is GA. What the production runtime promises for guardrails, observability, and scaling, and what you must verify yourself.
-
The Microsoft Agent Framework Claude Code Connector Is the Real Story (Copilot Ships Too)
October 28, 2026
Microsoft Agent Framework Claude Code and GitHub Copilot connectors just shipped. Why the Claude one is the strategy tell, and how to architect multi-model routing.
-
Microsoft Agent Framework DevUI: Inspectability Just Became the Price of Entry
October 26, 2026
Microsoft Agent Framework DevUI makes agent debugging a local inner-loop problem. What it does, what it misses, and how to slot it next to Foundry tracing.
-
Agents That Remember What Worked: The Promise and the Trap in Microsoft Agent Framework Memory
October 23, 2026
Microsoft Agent Framework memory now records and replays successful task procedures in public preview. Why it's the real unlock, and the governance trap inside.
-
Microsoft Agent Framework CodeAct: The Quiet Admission That Tool-Call Chains Were a Tax
October 19, 2026
Microsoft Agent Framework CodeAct collapses tool-call chains into one executable block, and Hyperlight makes it shippable. Sequential agents just became debt.
-
Azure Functions Agents Runtime: Stop Paying for Idle Agents
October 16, 2026
The Azure Functions agents runtime makes hosting cost proportional to agent activity. Why bursty agents belong on serverless, with the TCO math to prove it.
-
Microsoft Agent Framework Tracing Is Here - Stop Shipping Agents You Can't Debug
October 14, 2026
Microsoft Agent Framework tracing is in public preview for Python and .NET. What it captures, how OpenTelemetry fits, and why to wire it in on day one.
-
Foundry Hosted Agents in C#: What the One-Package, Two-Command Deploy Actually Hides
October 12, 2026
Foundry hosted agents in C# turn agent deployment into a boring app workload. What the published deploy path hides, where it breaks, and how to ship a pilot.
-
Microsoft Agent Framework Declarative Workflows 1.0: Stop Burying Orchestration in Your App Code
October 9, 2026
What Microsoft Agent Framework Declarative Workflows 1.0 moves into YAML, why that fixes agent orchestration governance, and when to adopt versus wait.
-
Microsoft Agent Framework New Features: Memory, Resilient Execution, and the New Bar for Production Agents
October 7, 2026
Microsoft Agent Framework's new memory, resilient execution, and interface features quietly redefine the minimum viable production agent. Audit yours against it.
-
Exam AI-500 Beta: Microsoft Just Published Its Multi-Agent Roadmap and Called It a Certification
July 23, 2026
Exam AI-500 beta is open. What the Multi-Agent AI Solutions Expert credential tests, how it differs from AI-102 and AB-100, and who should book versus wait.
-
Copilot Studio Skills Gallery: What an Offline Converter Signals
July 17, 2026
A fully offline document converter in the CAT skills gallery shows where Copilot Studio skills are heading, and what architects should decide this week.
-
Agent Skills for Python: Why It Ends the Mega-Prompt Era
July 15, 2026
Agent Skills for Python shipped stable in Microsoft Agent Framework. What that means for mega-prompt agents, Prompt Flow, and your token bill.
-
AB-100 Decoded (2026): What the Agentic AI Architect Exam Tests
July 11, 2026
AB-100 replaces four retired Microsoft business-apps certifications. What the agentic AI architect exam tests, its prerequisite gate, and how to prep.
-
Agent Framework 1.0 Orchestration Patterns: How to Choose the Right One
July 8, 2026
Microsoft Agent Framework 1.0 stabilizes five orchestration patterns in Python and .NET. How to pick one, where agents beat flows, and when a flow still wins.
-
Microsoft Foundry June 2026 Release: What It Really Decides for You
July 8, 2026
Foundry's June 2026 release makes agents cheap to ship. What actually shipped, and the decision rule for choosing a flow, a Foundry agent, or a custom Azure build.
-
Agent Skills for .NET Left Preview: What It Changes and What to Migrate First
July 7, 2026
Agent Skills for .NET is out of experimental preview. Where it moves the line between agents and flows, where it does not, and what to migrate first.
-
How to Become an Agent Product Owner: Skills and a 30-60-90 Plan
June 30, 2026
Agents fail on ownership, not model quality. The four jobs an agent product owner owns, the skill stack it takes, and a 30-60-90 path to staff or become one.
-
Claude Code in Azure ML Pipelines: Agentic CI/CD With an Audit Trail
June 30, 2026
Claude Code as an Azure ML pipeline step: tests as the control plane, MLflow lineage, risk-tiered gates, and a human checkpoint before production.
-
Stop Vibe-Coding Power Platform: Turn ADO Work Items Into Specs Any AI Agent Can Build From
June 30, 2026
Agentic development for Power Platform pays off only when the Azure DevOps work item is a machine-readable spec any agent can build from and verify against.
-
Microsoft Agent vs Flow: When to Build an AI Agent, When to Automate a Workflow (and When to Do Neither)
June 30, 2026
A determinism-first framework for the Copilot Studio vs Power Automate call, with a six-dimension rubric scoring Agent, Flow, Hybrid, or Neither.
-
Dataverse MCP Server: Tool Shape Is Your New Security Perimeter
June 24, 2026
Microsoft's Dataverse MCP Server turns your system of record into agent-callable tools. The contract you define, not the platform, decides what an LLM can do.
-
When Not to Build an AI Agent: A 5-Question Test (2026)
June 18, 2026
Before you build an AI agent, run five questions. When a flow, a query, or nothing at all beats an agent, with worked examples and the agent tax to budget.
-
Your AI Agent Project Is Really a Data Project: The Data-Prep Tax
June 18, 2026
AI agent projects are really data projects. Why data preparation and upkeep, not the model, decides whether an agent ships, and how to scope the data work first.
-
AI Cost Governance in 2026: The Spend Caps That Don't Actually Cap
June 17, 2026
AI cost governance in 2026: the budget you reach for first only alerts, it never stops. A vendor-by-vendor playbook on which AI spend controls hard-stop and which just notify.
-
Microsoft's Agents Hub Decoded (2026): The Frameworks and the Gaps
June 17, 2026
Microsoft's new Agents hub formalizes agent architecture, archetypes, a maturity model, and evaluation. An architect's read on what it still leaves to you.
-
Coding Agents and Developer Economics on the Microsoft Stack (2026)
June 14, 2026
Coding agents are normal tool evolution, not rupture. The Microsoft business-apps read: execution compresses, but judgment and accountability do not.
-
Rayfin Decoded: Microsoft's Bet on Prompt-to-Production Backends
June 13, 2026
Rayfin lets agents define and deploy a full backend to Microsoft Fabric in one command. What it actually compresses, what it couples you to, and when an architect should reach for it.
-
Foundry Hosted vs In-Process vs Copilot Studio Agents (2026 Decision)
May 31, 2026
Foundry Hosted vs in-process vs Copilot Studio agents: a 2026 four-gate decision framework that picks the right path by who builds it and who runs it.
-
Six Rules for LLM-Agnostic AI Agents on Microsoft Foundry
April 29, 2026
Foundry Model Router routes 18 models across providers. Six rules for AI agents that survive the next vendor rotation, with a proposal-writing example.