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
AI Agents and Agentic AI Architecture in Python · LearnSpace
Назад в каталог
courseraАнализ данных

AI Agents and Agentic AI Architecture in Python

Курс от Vanderbilt University
Начальный≈ 7.3 чАнглийский
О курсеНавыкиПрограммаПреподаватели

О курсе

Master the Art of Building Intelligent Python Agents That Think, Reason, and Act Unlock the full potential of Python for creating autonomous AI agents that solve complex problems without constant human direction. In this comprehensive course on AI Agents and Agentic AI with Python & Generative AI, you'll learn how to architect sophisticated agent systems that leverage Python's robust ecosystem and industry-standard capabilities. This course takes you beyond the foundations covered in the AI Agents and Agentic AI with Python & Generative AI course to explore advanced patterns for building truly intelligent agents in Python. You'll delve into specialized techniques like self-prompting, expert personas, document-as-implementation, and multi-agent orchestration - all implemented with Python's powerful frameworks and libraries. What You'll Learn: - Self-Prompting Patterns in Python: Build agents that dynamically adopt different thinking modes to handle specialized tasks, transforming unstructured data into structured formats with clean Python implementations - Python-Based Expert Persona Systems: Implement consultation frameworks where agents can invoke domain experts for specialized knowledge while maintaining clean architecture - Document-as-Implementation: Use Python's powerful file handling to create systems where human-readable documents become executable business logic - Multi-Agent Collaboration with Python: Design sophisticated memory sharing and coordination mechanisms between specialized Python agents - Progress Tracking & Planning: Implement robust planning and reflection capabilities using Python's comprehensive tooling - Python Agent Safety & Trust Systems: Build transaction management and safety mechanisms that leverage Python's exception handling and security features By the end of this course, you'll be equipped to build complex, production-ready agent systems in Python that can reason across multiple domains, handle complex workflows, and safely interact with real-world systems. Whether you're building productivity tools, automating complex business processes, or creating intelligent assistants, you'll have the Python-specific knowledge to implement agentic AI solutions that provide genuine business value. This course will teach you these concepts using OpenAI's APIs, which require paid access, but the principles and techniques can be adapted to other LLMs.

Навыки, которые вы освоите

Prompt EngineeringOpenAI APIFile I/OBusiness LogicTransaction ProcessingLLM ApplicationAgentic WorkflowsPrompt PatternsGenerative AI AgentsAI OrchestrationMemory ManagementAgentic systemsAI WorkflowsAI SecuritySoftware Architecture

Программа курса

5 модулей · 35 учебных материалов

01Extending AI Agents with Self-Prompting14 материалов
Prompts as ComputationВидеоSelf-Prompting & Clean Separation of AI Agent ReasoningPLUGINBridging Computer Tools & Unstructured Data with Prompting - the AI ShimВидеоAI Agent Structured Data ExtractionPLUGIN

Учитесь у экспертов

Dr. Jules White

Professor of Computer Science

AI Agents and Agentic AI Architecture in Python
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

≈ 7.3 ч

5 модулей

Язык: Английский

Субтитры: Арабский, Французский, Узбекский, Итальянский, Бразильский португальский, Корейский, Немецкий, Индонезийский, Испанский, Японский, Казахский, Венгерский

Часть программы вашего университета
An Invoice Processing AgentPLUGIN
The Persona Pattern and Reasoning - Personas are an efficient programming abstractionВидео
The Persona PatternВидео
Format of the Persona PatternЧтение
Simple Multi-Agent Systems with PersonasВидео
Consulting Experts or Simulating with the Persona PatternPLUGIN
The Persona Abstraction & AgentsPLUGIN
Invoice Processing with ExpertsPLUGIN
Using Human Policies for Document-as-ImplementationPLUGIN
Persona & Self-Prompting ReviewЗадание
02AI Agent Design Principles & Safety3 материалов
The MATE Design Principles for AI AgentsВидеоMATE Design Principles in CodePLUGINAI Agents & Environment SafetyPLUGIN
03Multi-Agent Systems8 материалов
Introduction to Multi-Agent SystemsВидеоBuilding Multi-Agent Systems: Agent-to-Agent CommunicationPLUGINAgent Interaction & MemoryВидеоAgent Interaction Patterns with MemoryPLUGINRemoving Noise: Focusing Agent AttentionВидеоAdvanced Agent InteractionPLUGINProviding Agentic AI Information About the WorldВидеоAgent Interaction ArchitecturesЗадание
04Dependency Injection for Tools3 материалов
Isolating Agents from Accidental ComplexityВидеоClean AI Tools with Dependency InjectionPLUGINClean Tool Dependency Injection with the EnvironmentPLUGIN
05Approaches to Improving AI Agent Reasoning7 материалов
Improving AI Agent Reasoning with In-Context LearningВидеоImproving AI Agent Reasoning with Up-front Planning & Chain of ThoughtВидеоThe Capability Architectural PatternPLUGINAhead of Time Planning for Improving Agent ReasoningPLUGINImproving AI Agent Reasoning with In-loop PlanningВидеоIntermediate Planning: Tracking Progress in the Agent LoopPLUGINThe Great Agent Trade-off: Ahead of Time vs. DynamicВидео