Teach on Udemy

Turn what you know into an opportunity and reach millions around the world.

Learn More

Your cart is empty.

Keep shopping

Agentic AI on AWS - LangGraph, CrewAI, Bedrock, Kiro & MCP

Name: Agentic AI on AWS - LangGraph, CrewAI, Bedrock, Kiro & MCP
Rating: 4.5 (2598 reviews)

Build 5 production-ready AI agents using LangGraph, CrewAI, Amazon Bedrock, Kiro and MCP

Created byRahul Trisal

Last updated 8/2026

English

German [Auto],English [Auto],

What you'll learn

Learn fundamentals about AI Agents such as Planning, Tools, Memory and Multi-Agent Framework
Project 1 – AnyCart Multi-Agent E-Commerce Application (LangGraph + Amazon Bedrock)
Project 2 – Vacation Planner Multi-Agent AI App (CrewAI + Amazon Bedrock)
Project 3 –Build FleetMate Application using AWS KIRO - Agentic AI powered Integrated Development Environment (IDE)
Project 4 - Hotel Booking Agent : Build a Hotel Booking Agent using Amazon Bedrock Agents (Depreciated for new users)
Project 5 - Build an Infra Coding Agent using AWS MCP CloudFormation Server and Amazon Q CLI.
Learn Basic Concepts of CrewAI
Learn Basic Concepts of LangGraph AI framework
Learn basics of Model Context Protocol (MCP) and build Coding Agent with AWS Cloudformation MCP Server and Amazon Q CLI
Detailed Refresher on Generative AI, Amazon Bedrock, Prompt Engineering and Knowledge Bases
Refresher on Basic Python, AWS Lambda and Boto3

Course content

13 sections • 100 lectures • 14h 7m total length

Course Introduction3:09

Shortcomings of the LLM based apps5:52
Overview of AI Agents8:35
AI Agents - Planning11:39
AI Agents - Planning (Hands-On)9:13
AI Agents - Tools and Actions6:24
AI Agents - Tools and Actions (Hands-On)5:46
AI Agents - Memory6:34
Agentic AI Framworks - Mental Model for building Agentic AI Apps on AWS6:07
Adopt a model for building agentic AI apps on AWS using LangGraph and KriviAI. Deploy with Lambda, ECS, EKS, or Bedrock Agent Core, and leverage tools, model context protocol.
Evolution of AI Agents and Core Capabilities

Business Use Case - AnyCart E-Commerce Agentic AI App4:24
Build an e-commerce Anycard multi-agent AI app with the Langraph AI framework on AWS to provide recommendations, place orders, and track orders via intent-driven routing.
Agentic Architecture using LangGraph - AnyCart E-Commerce6:04
Build a layered LangGraph architecture that routes shopper intents to recommendation, order placement, and order tracking using Bedrock LLM, prompts, and tools, with a product catalog in S3.
LangGraph Fundamentals : Graph, State, Agents, Nodes, Edges, Tools and Compile8:45
LangGraph provides an open source AI agent framework built on LANG chain to build, deploy, and manage agentic workflows, featuring nodes, agents, edges, tools, state, and a compiled graph.
Building Product Recommendation Backend (Amazon Bedrock Knowledge Bases)12:06
Build a product recommendation backend by uploading a product catalog PDF to S3, creating a Bedrock unstructured vector knowledge base, syncing data, and enabling retrieval-based answers for the agent.
Recommendation Tool (AWS Lambda)17:11
Create an AWS Lambda tool named recommend_product that uses boto3's Retrieve and Generate API to query the Bedrock knowledge base and return a generated answer.
Order Management Backend (Amazon DynamoDB)2:02
Provision a DynamoDB table named orders with order_id as the partition key, and deploy a place-orders Lambda to write orders from user requests via the orchestrator.
Order Management Tool (AWS Lambda)14:27
Learn how the place orders AWS Lambda function writes product id and quantity to the DynamoDB orders table, generates an order id, and sets status to pending for tracking.
Order Tracking Tool (AWS Lambda)13:12
Develop an order tracking tool by wiring an AWS Lambda function order_status to query the DynamoDB orders table with the incoming order_id using get_item and return the order details.
Designing the LangGraph State and Models9:32
Define the LangGraph state and models using Amazon Bedrock Converse in LangChain, configuring the Bedrock model, region, and imports, and establishing a shared state for orchestrator and recommendation agents.
Orchestrator Agent, Recommendation and Specialized Nodes18:07
Orchestrator agent routes user input to recommendation, order management, or order tracking nodes based on intent via a shared state and json outputs.
Building, Compiling, and Invoking the LangGraph Workflow13:04
Build, compile, and invoke a LangGraph workflow by defining nodes such as orchestrator and recommendation agent, wiring edges and conditional routes, memory saver, and testing with Python and AWS CLI.
Deploying the LangGraph Application on AWS Lambda13:27
Building the Streamlit Frontend5:23
AnyCart E-Commerce Application – End-to-End Demo2:25

CrewAI Introduction7:16
Explore how crew AI provides a lean python, cloud-agnostic multi-agent platform to build automated workflows with any llm, where agents handle tasks and process under a crew orchestrator.
CrewAI Detailed Concepts13:01
CrewAI - Pre Requisites3:20
CrewAI - Vacation Planner Implementation - 112:18
Set up a vacation planner app with Crew AI and Amazon Bedrock, generating boilerplate scaffolding and configuring agents and tasks. Review the folder structure, main files, and dot env setup.
CrewAI - Vacation Planner Implementation - 218:41
CrewAI - Vacation Planner Implementation - 310:58
CrewAI - UI Deployment8:23

Evolution of AI-Powered Software Development and Kiro4:10
Explore the evolution of AI-powered software development on AWS from code whisperer to Q developer to Kiro, and learn how Kiro enables spec-driven development by generating requirements, design, and code.
Introduction to Kiro4:55
Important - Kiro Architecture, and How It Works11:55
Kiro Installation and Setup (Free Tier)3:48
Kiro Enterprise Setup - IAM Identity Center7:23
Business Use Case — FleetMate4:33
Writing Effective Requirement.md Files with Kiro (Hands-On)11:44
Learn to craft requirements for a feature using Kiro, generating a requirements.md with user stories and acceptance criteria for an AI powered vehicle diagnostic assistant, including feedback-driven refinements.
Creating Design.md and Task.md Files with Kiro (Hands-On)8:31
How Kiro integrates with external services using MCP5:28
Integrating Jira and Confluence MCP with Kiro (Hands-On)19:17
Steering Files in Kiro - Overview6:15
Steering Files in FleetMate (Hands-On)5:44
Building FleetMate End-to-End14:03
Hooks in Kiro - Overview5:17
Hooks in Kiro (Hands-On)7:22
Kiro Powers Overview8:35
Kiro Powers (Hands-On)7:36
Kiro - Model capabilities and best practices7:05

Amazon Bedrock Agents - Overview12:46
Amazon Bedrock Agents - Console Walkthrough15:13
Amazon Bedrock Agents - How do Agents work ?5:26
Learn how Amazon Bedrock agents work by building a final prompt from user input, instructions, tools, and guardrails, with a session history guiding task decomposition and planning.
Important - Course Rating and Feedback0:17
Dear Learner,
Hope you are having a good experience.
Please do rate the course and write feedback if possible.
It helps me stay motivated to add more content and use cases to the course.
You can rate the course from 'Leave a Rating''section on TOP Right Corner as shown in the image.
If you face any issues with the course, please send me a mail on rahultrisal@gmail.com
Common Use Cases for AI Agents6:03
Quiz 2 : Amazon Bedrock Agents (Deep Dive)

Business Use Case - Hotel Booking Agent1:38
Use Case Solution Architecture5:48
Charges for Implementing the Use Case (if following along) and Troubleshooting3:35
Agent Creation9:23
Agent Integration with Knowledge Bases for Room Information12:53
Hotel Room Availability - Arch and DynamoDB creation5:19
Hotel Room Availability - Lambda Function Creation18:41
Hotel Room Availability - OpenAPI Schema and Integration with Agent15:03
Hotel Room Booking - DynamoDB Creation3:55
Hotel Room Booking - Lambda Creation : Part 111:20
Hotel Room Booking : Lambda Creation - Part 212:10
Hotel Room Booking - OpenAPI Schema Creation4:45
Hotel Room Booking - Agent & Lambda & Action Group Integration14:08
Agent Guardrails and Memory Setup7:12
Agent Version and Alias Deployment1:40
Deploy the hotel room booking agent on Amazon Bedrock, publish the baseline version, and create aliases and versions for dev, QA, and prod.
Streamlit UI Pre-Requisites7:41
Agent UI Deployment with Streamlit5:43
Final Demo2:15
Frontend Deployment to EC2 Server19:14
Adding Application Load Balancer to EC26:56
Front-end your UI on EC2 by using an application load balancer. Register a target group and route HTTP traffic to the Streamlit app on port 8501.
Important - Bedrock Knowledge Bases Cleanup1:31

Business Use Case2:24
Use Case Architecture3:25
Collaborator Agent 1 - HR Agent (Backend and Agent Creation)7:57
Collaborator Agent 1 - HR Agent (Lambda and Agent Integration)17:27
Collaborator Agent 1 - HR Agent (Bedrock KnowledgeBase)10:20
Collaborator Agent 2 - Hotel Room Booking Agent0:27
Supervisor Agent (Agent 3) - Enterprise Travel Agent12:55
UI using Streamlit for Enterprise Travel Agent (Multi-Agent)4:26
Final Demo - Enterprise Travel Agent (Multi-Agent)2:52

Requirements

Basic Knowledge of AWS Services. Knowledge of Amazon Bedrock and AWS Lambda will be nice to have but a refresher has been provided incase of lack of familarity of these services

Description

Build enterprise grade AI Agents on AWS with LangGraph, CrewAI, Amazon Bedrock, Kiro & AWS MCP | Hands-On

Welcome to one of the most comprehensive hands-on courses for building AI Agents on AWS using LangGraph, CrewAI, Amazon Bedrock, AWS MCP, Kiro, Amazon Q CLI, and Large Language Models (LLMs).

Whether you're an AWS Developer, Solution Architect, Software Engineer, or AI Engineer, this course will help you build production-style AI applications through real-world projects—not just theory.

The course is taught by a practicing AWS Solution Architect and best-selling Udemy Instructor with 17+ years of AWS experience, helping Fortune 500 customers build cloud and AI solutions.

What You'll Learn

You'll begin with the fundamentals of Agentic AI and gradually progress to building complete AI Agent applications.

Topics covered include:

What is Agentic AI?
AI Agent architecture and design patterns
Planning and reasoning, Memory and Tool Calling
LangGraph Agentic AI Framework
CrewAI AI Framework
Amazon Bedrock
Kiro
Q CLI

Every concept is reinforced with practical coding exercises and real-world use cases.

Hands-On Projects

Project 1 – AnyCart Multi-Agent E-Commerce Application (LangGraph + Amazon Bedrock)

Build a complete multi-agent e-commerce application using LangGraph.

You'll learn how to:

Design AI workflows using graphs
Build multiple collaborating agents
Integrate Amazon Bedrock
Implement recommendation tools
Build tool-calling agents
Orchestrate complex AI workflows

Project 2 – Vacation Planner Multi-Agent AI App (CrewAI + Amazon Bedrock)

Build an enterprise-style AI Vacation Planner using CrewAI.

Learn how multiple specialized AI agents collaborate to:

Research destinations
Create travel itineraries
Generate personalized recommendations
Produce professional travel reports

Project 3 – FleetMate using AWS Kiro

Build a real-world application using AWS Kiro, AWS's Agentic AI-powered IDE.

Learn how AI-assisted software development accelerates application delivery using specification-driven development.

Project 4 – Hotel Booking AI Agent (Amazon Bedrock Agents) -- > No longer available for new builders

Build a Hotel Booking Agent using:

Amazon Bedrock Agents
AWS Lambda
Amazon Bedrock Knowledge Bases
Amazon Nova Models

Important Note: Starting 1 August 2026, Amazon Bedrock Agents are no longer available for new customers. This project is retained for existing users who still have access and to help learners understand the evolution of AWS Agentic AI services.

Project 5 – Infrastructure Coding Agent using AWS MCP

Build an Infrastructure-as-Code AI Agent using:

AWS MCP CloudFormation Server
Amazon Q CLI
Model Context Protocol (MCP)

Learn how AI agents can generate, understand, and automate AWS infrastructure.

Included Refresher Modules

If you're new to AWS Generative AI, don't worry.

The course includes refresher modules on:

Generative AI fundamentals
Amazon Bedrock
AWS Lambda

These provide the foundation needed before building the hands-on projects.

Many learners ask which order they should take my AWS AI courses. Based on my experience working with enterprise customers, I recommend the following learning path.

Learning Path – Become an AWS Generative AI Developer or Architect

Hands-On Learning Path (Recommended)

These courses focus on building practical skills used in real-world projects.

1. Amazon Bedrock, Amazon Q & AWS Generative AI [Hands-On] (Beginner)

Start here if you're new to Generative AI and Amazon Bedrock.

You'll build a strong foundation before moving into AI Agents.

2. Build AI Agents on AWS with LangGraph, CrewAI, Amazon Bedrock, Kiro & AWS MCP (Intermediate)

This course focuses on designing and building Agentic AI applications using modern AWS services and frameworks.

Recommended after completing Course 1.

3. Amazon Bedrock AgentCore: Deploy AI Agents on AWS (Advanced)

Learn how to deploy, scale, secure, observe, and operate AI Agents in production using Amazon Bedrock AgentCore.

Recommended after completing Courses 1 and 2.

AWS Certification Learning Path

1. AWS Certified AI Practitioner (AIF-C01)

Recommended after completing Course 1 or alongside this course.

Outcome:

Understand AWS Generative AI services
Build confidence with AI concepts
Prepare for and pass the AWS Certified AI Practitioner (AIF-C01) certification

2. AWS Certified Generative AI Developer – Professional Certification Course

This advanced certification course will help you prepare for the AWS Certified Generative AI Developer – Professional exam through comprehensive theory, architecture discussions, and hands-on examples.

I look forward to helping you build real-world AI Agents on AWS and advance your career in Generative AI and Agentic AI.

Who this course is for:

Anyone looking to learn about AI Agents and build Agentic AI apps using AWS Amazon Bedrock Multi-Agent Framework, Kiro and CrewAI

Agentic AI on AWS - LangGraph, CrewAI, Bedrock, Kiro & MCP

What you'll learn

Explore related topics

Course content

Course Introduction1 lecture • 3min

Evolution of AI Agents and Core Capabilities8 lectures • 1hr

Build AnyCart Multi-Agent E-Commerce App using LangGraph on AWS14 lectures • 2hr 20min

GenAI AI Architect Roadmap on AWS: Skills You Need to Learn in 2026 (Optional)1 lecture • 10min

Crew AI - Build Vacation Planner Agent with Crew AI7 lectures • 1hr 14min

Mastering Kiro — Spec-Driven AI Software Development18 lectures • 2hr 24min

Amazon Bedrock Agents - Deep Dive5 lectures • 40min

Hotel Booking Agent21 lectures • 2hr 51min

Enterprise Travel Agent (Multi-Agent)9 lectures • 1hr 2min

Model Context Protocol (MCP) : Build Infra Agent with AWS MCP5 lectures • 45min

Requirements

Description

Who this course is for: