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
Building AI Research Agents: LangChain, RAG, and LangGraph · LearnSpace
Назад в каталог
courseraАнализ данных

Building AI Research Agents: LangChain, RAG, and LangGraph

Курс от Board Infinity
Средний≈ 14.8 чАнглийский
О курсеНавыкиПрограммаПреподаватели

О курсе

This fully hands-on, project-based course walks you through building a real-world, production-ready AI-powered Smart Research and Writing Assistant entirely from scratch using LangChain and OpenAI. Every video is code-along, and every concept is applied immediately to one continuous project that grows in complexity across all four modules — mirroring how real AI agents are actually built in production environments. You'll start by setting up a professional development environment and building a foundational chatbot that connects to the OpenAI API, processes queries through prompt templates, and returns structured responses using Pydantic. Next, you'll teach the assistant to act in the real world by integrating Tavily web search, building custom tools with the @tool decorator, StructuredTool, and BaseTool, and implementing the ReAct reasoning pattern so the agent autonomously decides when to search, calculate, or respond. You'll then make it context-aware by adding memory strategies, migrating to LangGraph for full state management, conditional routing, checkpointing, and human-in-the-loop approval for sensitive actions. In the final module, you'll bring the assistant to production quality: adding Retrieval-Augmented Generation (RAG) over uploaded documents with ChromaDB, splitting the system into coordinated Researcher, Writer, and Reviewer sub-agents, building a Streamlit front-end, and deploying the complete containerized system with LangSmith monitoring and safety guardrails. By the end, you'll have a fully functional, portfolio-ready multi-agent assistant you can showcase to employers. The course spans ~20 hours across 4 modules, 12 lessons, and 48 videos, with roughly 5.5 hours of video content. Independent Course Disclaimer Disclaimer: This is an independent educational resource created by Board Infinity for informational and educational purposes only. This course is not affiliated with, endorsed by, sponsored by, or officially associated with any company, organization, or certification body unless explicitly stated. The content provided is based on industry knowledge and best practices but does not constitute official training material for any specific employer or certification program. All company names, trademarks, service marks, and logos referenced are the property of their respective owners and are used solely for educational identification and comparison purposes.

Навыки, которые вы освоите

Prompt EngineeringAI OrchestrationLLM ApplicationTool CallingAgentic systemsLarge Language ModelingAI WorkflowsRetrieval-Augmented GenerationGenerative AI Agents

Программа курса

4 модулей · 59 учебных материалов

01Foundations: Setting Up the Research Assistant15 материалов

Connecting to OpenAI and Building the First Response

Understanding Messages, Roles, and Parameters in the Chat APIВидеоInitialising LangChain ChatOpenAI and Running a Research QueryВидеоHandling API Errors, Rate Limits, and Response ValidationВидеоIntroduction to LLMs, RAG & Project SetupЗадание

Prompt Engineering and Structured Outputs for Research

Учитесь у экспертов

Board Infinity

Instructor

Building AI Research Agents: LangChain, RAG, and LangGraph
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

≈ 14.8 ч

4 модулей

Язык: Английский

Часть программы вашего университета
Configuring the OpenAI API Key and Making the First API CallВидео
Understanding Messages, Roles, and Parameters in the Chat APIВидео
Initialising LangChain ChatOpenAI and Running a Research QueryВидео
Handling API Errors, Rate Limits, and Response ValidationВидео
Document Ingestion & ParsingЗадание

Chunking, Tokenization & Text Structuring

Designing the Research Assistant System Prompt and PersonaВидеоBuilding Reusable Prompt Templates with LangChainВидеоDefining Structured Output Schemas with PydanticВидеоPutting It Together: End-to-End Research Query to Structured OutputВидеоChunking, Tokenization & Text StructuringЗаданиеFoundations, Document Ingestion & Text ProcessingЗадание
02Tools and Function Calling: Teaching the Assistant to Search16 материалов

Embeddings & Vector Databases

How Function Calling Works: From User Query to Tool ExecutionВидеоOpenAI Tool Schemas: Defining Parameters the Agent Can UseВидеоHands-On: Adding Tavily Web Search as the First Tool ВидеоTesting the Search Tool: Running Live Research QueriesВидеоEmbeddings & Vector DatabasesЗадание

Prompt Engineering & LLM Orchestration

Creating a Source Formatter Tool with the @tool DecoratorВидеоBuilding a Summariser Tool with StructuredTool and Pydantic SchemasВидеоBuilding a Citation Extractor Tool with BaseTool for Complex LogicВидеоTool Validation, Error Handling, and Unit Testing Your ToolsВидеоPrompt Engineering & LLM OrchestrationЗадание

Document Summarization Engine

The ReAct Pattern: How the Agent Reasons Before ActingВидеоAssembling the Agent: create_react_agent() with All ToolsВидеоRunning Multi-Step Research: The Agent Searches, Formats, and Responds ВидеоDebugging Agent Decisions: Tracing with LangSmithВидеоDocument Summarization EngineЗаданиеEmbeddings, Vector Search & SummarizationЗадание
03Retrieval-Augmented Generation & Q&A16 материалов

Building the Core RAG Q&A Pipeline

Why the Assistant Forgets: Understanding Stateless vs Stateful AgentsВидеоImplementing ConversationBufferMemory for Full History Recall ВидеоOptimising with Summary and Window Memory for Long SessionsВидео Persisting Conversation History to File and Database StorageВидеоGeospatial Map VisualizationsЗадание

Multi-Chart Layouts & Interactivity

Why LangGraph? From AgentExecutor to Graph-Based OrchestrationВидео Defining the Research Assistant State with TypedDictВидеоBuilding Nodes: LLM Node, Search Node, and Format NodeВидеоAdding Conditional Edges: Should the Agent Search or Respond Directly?ВидеоMulti-Chart Layouts & InteractivityЗадание

The Art of Visual Storytelling

Checkpointing: Saving and Restoring Research Sessions with MemorySaverВидеоStreaming Agent Responses and Intermediate Research Steps to the UserВидеоHuman-in-the-Loop: Adding an Approval Step Before Report FinalisationВидеоEnd-to-End Test: Multi-Turn Research Session with Persistence and Approval ВидеоThe Art of Visual StorytellingЗаданиеAdvanced Visuals & Storytelling TechniquesЗадание
04Optimization, Deployment & Best Practices12 материалов

Dashboard Fundamentals with Dash

What Is RAG and Why Does the Research Assistant Need It?ВидеоLayouts with Dash Bootstrap ComponentsВидеоAdding KPI Cards & Static ChartsВидеоAgentic RAG: The Agent Decides When to Search Web vs Search DocumentsВидеоDashboard Fundamentals with DashЗадание

Interactivity with Callbacks

Building the Researcher Agent: Searches Web and DocumentsВидеоBuilding the Writer Agent: Produces Structured Research ReportsВидеоBuilding the Reviewer Agent: Quality Checks and Feedback Loop ВидеоInteractivity with CallbacksЗадание

Polishing, Deployment & Best Practices

Building the Streamlit Chat Interface for the Research AssistantВидеоPolishing, Deployment & Best PracticesЗаданиеBuilding & Deploying an Interactive DashboardЗадание