Design production-grade stateful agentic loops with Microsoft Foundry Agent Service. Implement run-status handling and context accumulation, configure agent reflection and planning cycles, architect session state persistence, build fork-based session patterns, and migrate existing Agents v1 workloads to the Agents v2 Responses API.
By the end of this module, you'll be able to:
- Design agentic loop patterns that handle run-status signals and accumulate context across iterations
- Implement agent reflection and planning cycles for multi-step reasoning
- Architect session state management strategies for persistent multi-turn agent interactions
- Build fork-based session patterns to support conversation branching and resumption
- Describe the Agents v2 runtime model including agents, conversations, responses, and items
- Migrate stateful agentic loop code from Agents v1 to Agents v2 using the azure-ai-projects 2.x SDK
Implement advanced multi-agent orchestration patterns in Microsoft Foundry. Design hub-and-spoke coordination with central orchestrator agents, implement parallel agent spawning with synchronization, configure supervisor patterns for hierarchical task delegation, and evaluate orchestration framework trade-offs.
By the end of this module, you'll be able to:
- Differentiate agentic AI from multi-agent AI and decide when multi-agent architectures earn their coordination cost
- Design hub-and-spoke multi-agent architectures with a central orchestrator agent
- Implement parallel agent spawning and synchronization patterns for concurrent execution
- Configure supervisor and coordinator patterns for hierarchical multi-level task delegation
- Evaluate orchestration framework trade-offs to select the right approach for a given scenario
Apply advanced task decomposition strategies in Microsoft Foundry multi-agent systems. Design prompt chaining workflows, implement LLM-driven adaptive decomposition, configure agent handoff protocols for context-preserving transitions, and balance decomposition granularity against coordination overhead.
By the end of this module, you're able to:
- Design prompt chaining workflows that guide agents through complex multi-step analytical tasks
- Implement dynamic adaptive task decomposition using LLM-driven meta-agent planning
- Configure agent collaboration and handoff protocols for context-preserving task transitions
- Balance task decomposition granularity against coordination overhead for performance optimization
Design enterprise-scale agent communication architectures using the A2A protocol in Azure. Implement agent discovery registries with capability-based routing, design shared state management across distributed agent networks, configure context isolation for multitenant deployments, and build conflict detection and resolution mechanisms.
By the end of this module, you're able to:
- Design A2A agent ecosystems with discovery registries and dynamic capability-based routing
- Implement shared state management strategies across distributed agent teams using Azure services
- Configure context isolation and sharing patterns for multitenant agent deployments
- Build conflict detection and resolution mechanisms for collaborative agent networks

Syllabus

Design stateful agentic loops with Microsoft Foundry Agent Service
- Introduction
- Examine production agentic loop architecture
- Examine the Foundry Responses API and Agents v2 model
- Implement agent reflection and planning cycles
- Design session state and context management
- Implement fork-based sessions and conversation resumption
- Migrate stateful agentic loops from Agents v1 to Agents v2
- Module assessment
- Summary
Implement advanced multi-agent orchestration patterns in Microsoft Foundry
- Introduction
- Differentiate agentic AI from multi-agent AI architectures
- Examine advanced orchestration architectures
- Implement hub-and-spoke orchestration
- Design parallel agent spawning and synchronization
- Compare orchestration frameworks
- Module assessment
- Summary
Apply task decomposition and agent collaboration strategies in Microsoft Foundry
- Introduction
- Design prompt chaining workflows
- Implement dynamic adaptive task decomposition
- Design agent handoff message schemas
- Ensure handoff reliability and context preservation
- Optimize decomposition granularity
- Module assessment
- Summary
Design enterprise-scale agent communication with A2A in Azure
- Introduction
- Design A2A agent ecosystems at scale
- Implement distributed shared state management
- Design context isolation and sharing strategies
- Build conflict detection and resolution mechanisms
- Resolve conflicts and maintain audit trails
- Module assessment
- Summary