MicrosoftLive sourceEvidence: Low35/100

AI agents and multi-agent systems invigorate enterprise technology at Microsoft Build 2025

Microsoft Build 2025 introduced significant advances in AI agents and multi-agent systems. Copilot Studio now lets organizations build multi-agent solutions where agents delegate tasks between each other, leveraging tools like Microsoft 365 Agent Builder, Azure AI Agents Service, and Azure Fabric. GitHub Copilot evolves into GitHub Coding Agent, expanding from pair programming to peer programming, while Copilot Studio now targets professional developers and supports external model integration from Azure Foundry. The Model Context Protocol (MCP) enables unified orchestration for agents across Microsoft’s stack. New local deployment features put AI Foundry (formerly Windows Copilot Runtime) on both Windows and Mac. SQL Server 2025 now features vector and AI search capabilities. These innovations establish robust foundations for generative, multi-agent, and autonomous business process automation across enterprise IT.

Organization
GitHub
Industry
Tech & Comms
Published
May 2025

Reported outcomes

Strategic outcomes

New product / capabilityEnabled multi-agent enterprise workflowsNew product / capabilityExpanded coding agent capabilitiesEcosystem & partnershipsUnified agent interoperability across platformsMarket & geographic expansionExtended AI deployments to Mac and Windows
Why do we believe this deployment?Customer identity, provider attribution, maturity, and source checks
Customer
GitHub
Provider
Microsoft
Maturity
Unknown

No explicit deployment-stage evidence found.

Customer identity supportedSource describes one deploymentMaturity evidence evaluated

Primary read

Use case focus

Showing 3 of 4

  • 1Multi-agent system
  • 2Peer programming
  • 3AI automation
  • Enterprises need autonomous multi-agent systems to handle complex workflows.
  • Development tools lack standardized, interoperable protocols for agent integration.
  • Organizations struggle to scale AI-driven business process automation securely and efficiently.
  • Need for advanced AI to work both in the cloud and at the edge.
  • Copilot Studio enables building multi-agent systems with Microsoft 365 Agent Builder and Azure AI Agents Service.
  • GitHub Copilot now acts as a coding agent for peer programming, automating advanced developer tasks.
  • Model Context Protocol (MCP) implemented across Microsoft apps for agent and model interoperability.
  • Local Windows AI Foundry and deployments expand platform reach to Windows and Mac.
  • SQL Server 2025 integrates AI and vector search for intelligent data capability.
  • Flexible enterprise agent deployments supporting cross-department use.
  • Unified AI protocol streamlines model and agent integration across cloud and on-premises systems.
  • Empowers professional developers with more powerful agent building tools.
  • Ecosystem-wide ability to leverage multi-agent systems for business process automation.
Architecture

Agents can be designed and orchestrated in Copilot Studio using Microsoft 365 Agent Builder, integrated with Azure AI Agents Service and Microsoft Fabric. MCP (Model Context Protocol) facilitates model and agent interoperability, with deployment options both on cloud and locally (AI Foundry). GitHub Coding Agent integrates with development platforms for codebase automation.

Sources & evidence1
Evidence: Low35/100Evidence strength
  • Customer explicitly identified
  • Technical implementation details available
  • Recent evidence check available
  • Last evidence check: Jul 22, 2026.
Live sourceStill referenced

The case's original source is still reachable.

  • Cited source last checked Jun 12, 2026 — ok (0/1 broken).

Measures whether this deployment's public evidence persists — not whether the system is still in production.

Published: May 19, 2025Publisher: Directions on Microsoft

AI-generated summary. Verify important details with the linked sources before relying on this case.

Explore related AI use cases

Was this useful?

Community

Comments

No published comments yet.