К содержимому
learnspaceYOUR NEXT CHAPTER
ПРОСТРАНСТВО ОБУЧЕНИЯ
ГлавнаяКаталог курсовМоё обучениеCoursera

Знания без границ

Учитесь у лучших университетов и компаний мира.

Открыть Coursera
Интеграция
Пространство университета
Моё пространствоСтраница курса
↵
ЯЛичный кабинетСтудент
© 2026 LearnSpaceКаждый день — возможность узнать больше.Помощь
Agentic AI Architectures with Patterns, Frameworks and MCP · LearnSpace
Назад в каталог
courseraПрограммирование

Agentic AI Architectures with Patterns, Frameworks and MCP

Курс от Packt
Средний≈ 13.6 чАнглийский
О курсеНавыкиПрограммаПреподаватели

О курсе

This course features Coursera Coach! A smarter way to learn with interactive, real-time conversations that help you test your knowledge, challenge assumptions, and deepen your understanding as you progress through the course. Unlock the next level of AI mastery by learning how to design and implement agentic AI architectures. This course empowers you to build intelligent multi-agent systems, understand core components like perception, planning, memory, and communication, and leverage frameworks and design patterns to create robust AI solutions. Your journey begins with a deep dive into AI agents, exploring principles, generative AI integration, and real-world use cases. You'll progressively design simple agents, then advance to sophisticated architectures, including orchestration, choreography, and multi-agent systems. Along the way, you'll learn the intricacies of Agentic RAG, communication protocols, and context engineering. The course then guides you through frameworks, patterns, and Model-Context Protocols (MCP), equipping you with hands-on skills to design AI workflows, agent orchestration, and platform integration. Practical projects such as building research assistants, ChatGPT agents, and health agents ensure that theory translates into applied expertise. This course is ideal for AI developers, software architects, and technical professionals aiming to implement agentic AI systems. No prior experience with MCP is required, but familiarity with AI fundamentals and microservices is recommended. Difficulty level: Intermediate to Advanced. By the end of the course, you will be able to design and implement agentic AI architectures, apply design patterns effectively, build multi-agent workflows, utilize Model-Context Protocols, and develop context-aware AI solutions across real-world applications.

Навыки, которые вы освоите

Agentic systemsModel Context ProtocolAgentic WorkflowsLangChainSoftware Design PatternsContext EngineeringContext ManagementRetrieval-Augmented GenerationGenerative AI AgentsAI IntegrationsTool CallingSoftware ArchitectureLLM ApplicationChatGPTInteroperabilityDeep LearningPrompt EngineeringAI Orchestration

Программа курса

13 модулей · 107 учебных материалов

01Introduction5 материалов

Introduction

IntroductionВидеоFull Course ResourcesЧтениеTools and Resources for the Course - Course SlidesВидеоCourse Project: Design Agentic AI Architecture for EShop MicroservicesВидео

Учитесь у экспертов

Packt - Course Instructors

Преподаватель курса

Agentic AI Architectures with Patterns, Frameworks and MCP
В каталоге вашей программы

Инвестируйте в себя

Новые знания — в удобное для вас время.

Начать на Coursera

Обучение откроется на Coursera
в новой вкладке

Обучение на Coursera

≈ 13.6 ч

13 модулей

Язык: Английский

Часть программы вашего университета
The Evolution: From Generative AI to Agentic AIВидео
02What are AI Agents? Key Principles8 материалов

What are AI Agents? Key Principles

What are AI Agents? Key PrinciplesВидеоHow AI Agents Use Generative AI?ВидеоWhen to Use AI Agents (And When Not To)ВидеоWhere to Use Agents? Real-world Use Cases for AI AgentsВидеоThe Hard Parts: Challenges and Limitations of AI AgentsВидео[DESIGN] Designing Your First Simple AI AgentВидеоUnderstanding AI Agents: Principles, Use Cases, and ChallengesDIALOGUEWhat are AI Agents? Key Principles - AssessmentЗадание
03Agent Anatomy - A Deep Dive Into Core Components14 материалов

Agent Anatomy - A Deep Dive Into Core Components

Introduction - Agent Anatomy: A Deep Dive Into Core ComponentsВидеоThe "Brain" - The Foundation Model (LLM)ВидеоThe "Senses" - Perception and Input HandlingВидеоThe "Blueprint" - Planning and Task DecompositionВидеоThe "Notebook" - The Architecture of MemoryВидеоThe "Hands" - Action and Tool CallingВидеоThe "Feedback Loop" - Learning and AdaptationВидеоThe "Voice" - Agent CommunicationВидео[DESIGN] Building a "Research Assistant" AgentВидеоThe Agentic Loop: Perception, Reasoning, Action, Learning (PRAL)Видео[DESIGN] Build ChatGPT Agent with The Agentic Loop (PRAL)Видео[DESIGN] Build Health Agent with The Agentic Loop (PRAL) & Human-in-the-loopВидеоDesigning the Agentic Loop: Anatomy of an AI AgentDIALOGUEAgent Anatomy - A Deep Dive Into Core Components - AssessmentЗадание
04What is Agentic AI?8 материалов

What is Agentic AI?

Introduction - What is Agentic AI?ВидеоAI Agent vs. Agentic AI: The Critical DifferenceВидеоWhat is Agentic Architecture? (Multi-Agent System (MAS))ВидеоHow Agentic AI Works: System-Level Architectures (Orchestration or Choreography)Видео[DESIGN] Orchestration and Choreography of Agentic AI ArchitecturesВидеоThe Hard Parts: Challenges of Agentic AI SystemsВидеоUnderstanding Agentic AI and Multi-Agent ArchitecturesDIALOGUEWhat is Agentic AI? - AssessmentЗадание
05From Traditional to Agentic Workflows7 материалов

From Traditional to Agentic Workflows

Introduction - From Traditional to Agentic WorkflowsВидеоComponents of Agentic Workflow: State, Nodes and EdgesВидеоAgentic Orchestration Patterns (Sequential, Concurrent, Group chat, Handoff)ВидеоFrameworks for Building Workflows (Code-First)ВидеоPlatforms for Building Workflows (Low-Code)ВидеоExploring Agentic Workloads and Workflow PatternsDIALOGUEFrom Traditional to Agentic Workflows - AssessmentЗадание
06The Frameworks Landscape7 материалов

The Frameworks Landscape

Introduction - The Frameworks LandscapeВидеоThe "Generalist Toolkit": LangChain & LangGraphВидеоThe "Data Specialist": LlamaIndex Deep DiveВидеоThe "Multi-Agent Specialist": Microsoft Agent FrameworkВидеоThe "Managed Platforms": Cloud Agents (AWS, Azure, Google)ВидеоComparing AI Agent Frameworks: When to Build, When to BuyDIALOGUEThe Frameworks Landscape - AssessmentЗадание
07The "Why" of Agentic Design Patterns16 материалов

The "Why" of Agentic Design Patterns

Introduction - The "Why" of Agentic Design PatternsВидеоThe Agentic Design PrinciplesВидеоThe "Tool Use" PatternВидеоThe "Planning" PatternВидеоThe "Reflection" Pattern (Metacognition)ВидеоThe "ReAct" Compound PatternВидеоIntroduction to Multi-Agent PatternsВидеоThe "Router & Specialist" PatternВидеоThe "Handoff" Pattern (Sequential Workflow)ВидеоThe "Group Chat / Debate" PatternВидеоThe "Swarm" Pattern (Parallelization)ВидеоIntroduction to Control & Safety PatternsВидеоThe "Human-in-the-Loop" (HITL) PatternВидеоThe "Ejection / Custom Logic" PatternВидеоExploring Agentic Design Patterns in AI SystemsDIALOGUEThe "Why" of Agentic Design Patterns - AssessmentЗадание
08The Limitations of Traditional RAG7 материалов

The Limitations of Traditional RAG

Introduction - The Limitations of Traditional RAGВидеоWhat is Agentic RAG?ВидеоAgentic RAG vs. Traditional RAG: The Critical DifferenceВидеоHow Agentic RAG Works: The "Active Researcher" LoopВидеоAgentic RAG Use CasesВидеоComparing Traditional RAG and Agentic RAG ArchitecturesDIALOGUEThe Limitations of Traditional RAG - AssessmentЗадание
09Agent Communication Protocols8 материалов

Agent Communication Protocols

Introduction - The "Protocol Problem"ВидеоWhat is the Agent Communication Protocol (ACP)?ВидеоWhat is the A2A (Agent-to-Agent) Protocol?ВидеоWhat is the Agent-user interaction (AG-UI) protocol?ВидеоWhat is the Model-Context Protocol (MCP)?ВидеоThe Architect's View: MCP vs. ACP vs. A2AВидеоUnderstanding Agent Communication Protocols and Their Use CasesDIALOGUEAgent Communication Protocols - AssessmentЗадание
10Model-Context Protocol (MCP)8 материалов

Model-Context Protocol (MCP)

Introduction - Why MCP?ВидеоThe MCP Architecture: MCP Host/Client, Server and ProtocolВидеоMCP Core Concepts & Definitions (The "Spec" Deep Dive)ВидеоThe MCP Lifecycle (How it Works)ВидеоThe Architect's View: MCP vs. OpenAPI (The "Why")ВидеоMCP Design Patterns & Case StudiesВидеоUnderstanding MCP Architecture and Design PatternsDIALOGUEModel-Context Protocol (MCP) - AssessmentЗадание
11Context Engineering7 материалов

Context Engineering

What is Context Engineering?ВидеоPrompt Engineering vs. Context Engineering (The Critical Shift)ВидеоThe Anatomy of Context (Types of Context)ВидеоStrategies for Effective Context EngineeringВидеоExample Use Cases of Context EngineeringВидеоArchitecting for AI: Fundamentals of Context EngineeringDIALOGUEContext Engineering - AssessmentЗадание
12AI Agent Development Lifecycle9 материалов

AI Agent Development Lifecycle

What is AI Agent Development? The Probabilistic ShiftВидеоThe Agent Development Lifecycle (ADLC): A 7-Step GuideВидеоDeep Dive: The Challenge of "Evaluation"ВидеоDeep Dive: "AgentOps" (Observability & Monitoring)ВидеоEnterprise Considerations: LLM Guardrails for Data Leakage, Prompt InjectionВидеоThe "Tool" Lifecycle: Managing MCP ServersВидеоThe "Blueprint": Reference Architecture for an Agentic AI PlatformВидеоNavigating the Agent Development Lifecycle (ADLC) in Enterprise AIDIALOGUEAI Agent Development Lifecycle - AssessmentЗадание
13Conclusion3 материалов

Conclusion

ThanksВидеоFull Course Practice AssessmentЗаданиеFull Course AssessmentЗадание