Teach on Udemy

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

Learn More

Your cart is empty.

Keep shopping

Master Google ADK - Agent Development Kit (Mac, Win, Ubuntu)

Name: Master Google ADK - Agent Development Kit (Mac, Win, Ubuntu)
Rating: 4.4 (84 reviews)

New ADK 2.0 | Build Multi Agent Systems | For Mac, Windows, Ubuntu | Deploy to Google Cloud | Free Gemini Key

Bestseller

Created byKartik Marwah

Last updated 5/2026

English

What you'll learn

We’ll start with a brief theory of agent frameworks, LLMs, and multi-agent systems.
We'll dive deep into Google's Agent Development Kit. All platforms (MacOS, Windows, Ubuntu) will be covered
We will build a practical "Website Builder Agent" from scratch.
And we'll do all of this using a free Gemini API Key, so you don't have to pay for AI models while you're learning.
We'll Integrate MCP using MCPToolset and Advanced tool use with Anthropic's Programmatic Tool Calling
We'll also cover the no-code Visual Agent Builder and Agent Config feature briefly

Course content

13 sections • 128 lectures • 6h 57m total length

About the Course & Instructor1:33
Introduction & What are Agent Frameworks?2:26
Why ADK? How is ADK Different?2:07
ADK UI Playground0:42
ADK Interoperability0:48
Summary, Important Points to Note!1:06

Introduction and Quick Preview of What We'll Build3:27
How to Build This Agent0:26
[macOS] Install Python, UV, VS Code and Git5:00
[Windows] Install Python, UV, VS Code and Git7:15
[Ubuntu] Install Python, UV, VS Code and Git4:42
How to Access the Course Code0:33
[macOS] Setup Code and Run5:23
[Windows] Setup Code and Run4:31
[Ubuntu] Setup Code and Run5:21
[macOS, Windows, Ubuntu] Setup Code and Run2:06
[macOS, Windows, Ubuntu] Code walkthrough8:54

Introduction and Overall Steps2:33
STEP 1 - Define Agent Architecture5:29
STEP 2 - Setup Folders6:01
STEP 3.1 - Coding the Requirements Writer Agent | Types of ADK Agents15:03
STEP 3.2 - Coding the Designer Agent8:03
STEP 3.3 - Coding the Code Writer Agent8:39
STEP 3.4 - Coding the Root Agent (Sequential Workflow Agent)5:13
STEP 3.5 - Environment Setup2:40
STEP 4 - Testing the Agent5:12
Test the ADK interface locally, select the Root Website Builder, and iterate design outputs—from requirements to design to HTML, CSS, and JS, validating the agent's responses.
Let's look at the Agent Output!2:28
Summary1:35
Demonstrates building an ai agent from sequential sub-agents that transform a high-level user query into detailed requirements, a design system, and a single html, css, and js page.

Parallel Agent Architecture and Code Structure7:38
Shows how to build a multi-agent system with Google's agent development kit, integrating a parallel workflow agent into a sequential route website builder to generate a single research web page.
Root Sequential Agent2:17
Questions Generator LLM Agent4:31
Questions Researcher Parallel Agent6:33
Query Generator LLM Agent3:13
Review of Agents from previous lectures3:44
Agent Setup and Demo7:24

Introduction and Agent Architecture4:13
Getting the code0:44
Clone or update your code from the adk_samples Repository0:39
Uvicorn server and fast api app6:30
Dockerfile5:32
Build a containerized multi-agent system for google cloud run with a python 3.13 slim dockerfile, /app as workdir, and requirements.txt; run uvicorn for the fastapi app on 0.0.0.0.
CLOUD SETUP - Create a new Gmail Account if you prefer2:35
CLOUD SETUP - Create a Google Cloud Project2:48
CLOUD SETUP - Install gcloud CLI5:33
CLOUD SETUP - Enabling API Keys if needed0:40
Environment Variables File4:29
Deploy to Google Cloud Run3:59
Demo of Agent on Google Cloud Run2:08
Demonstrates deploying the ADK agent on Google Cloud Run through the development UI. It generates a research report covering core components, architectural principles, multi-agent system facilitation, and tool integration.
Cleanup (important to avoid charges) and conclusion2:24

Quick Preview of the four methods3:44
Method 1: ADK WEB4:32
Method 2: ADK API SERVER4:23
Method 3 (Part 1): ADK Runner Program - Code Overview9:35
Method 3 (Part 2): ADK Runner Program - Agent Execution3:58
Method 4: ADK RUN Command Line Interface3:16
Demonstrates the ADK run command line interface by activating the virtual environment and generating a login page with Google sign-in and OTP auth.
Summary1:03

Why Build MCP Clients with Google ADK?1:05
STEP 1 - Building MCP Servers for our ADK based MCP Client12:02
Explore step 1 of building mcp servers for an adk based mcp client using the new streamable http transport, including a stateless server with add, subtract, multiply, and divide tools.
STEP 2 Building the ADK MCP Client0:09
Introduction, Preview of What We'll Build6:17
Code overview2:49
GOOGLE_API_KEY setup0:05
MCP Client User Interface - cmd.py4:30
MCP Client Implementation - client.py5:11
MCP Agent Implementation - agent.py8:33
Code - MCP config json file2:04
Code for Utilities & STDIO Server + Environment Setup8:17

Before we start - an important note for this section0:25
Introduction1:24
Quick Demo - What We’ll Build3:00
Getting the Code0:29
Visit the site to download the code or export it to GitHub. Then set up a Gemini API key to iterate with the AI assistant or build MCP server client.
Getting the Code0:10
Code Setup2:24
MCP Client Code Walkthrough - manager.py3:48
MCP Client - Server Discovery from Config1:34
MCP Client - Server Connection Logic3:13
Connect to all configured MCP servers by launching server processes, establishing json rpc sessions, and initializing client protocol sessions with automatic cleanup via exit stack.
MCP Client - Listing All Tools1:12
MCP Client - Calling a Tool1:42
MCP Client - Session Cleanup1:04
MCP Server - Simple Echo Server Example1:43
Explore a simple echo server built with fast MCP and the MCP tool decorator, showing an echo tool returning text and logs with v1 compatibility.
Running the MCP Client Manager0:59
ADK + MCP Integration - Overview and Step 1 (Load Server Config)3:06
ADK Integration - Step 2 (Connect to MCP Servers)3:07
Connect to MCP servers using the Google ADK MCP tool sets to manage connections automatically, then discover tools with toolset.getTools() and display them in a table with names and descriptions.
ADK Integration - Step 3 (Initialize Agent with McpToolset)0:50
ADK Integration - Step 4 (Initialize the Runner)0:42
Initialize an in-memory runner that drives the agent loop, taking user input, handling tool calls, executing tools, and returning results to the agent in a continuous query and response cycle.
ADK Integration - Step 5 (Start a Conversation Session)1:24
Start a conversation session with the runner's session service to let the agent remember context across messages, by providing an app name and user ID.
ADK Integration - Step 6 (Interactive Chat Loop)4:40
MCP Client Manager vs ADK Command Line Integration2:26
Running the Code1:18

Introduction to No-code Agents and YAML Config2:23
Setup your environment1:25
ADK Create & Get API Key2:23
Agent 1 - root agent2:44
Mult-Agent Orchestration1:48
Define a root agent with sub agents by listing their configuration files, enabling multi-agent orchestration that delegates Python tutoring or physics tutoring tasks and returns a final response.
Sub Agents - physics tutor1:29
Sub Agent - python tutor0:37
Putting it all together0:36
Running The Agent4:44

Requirements

You need either a MacOS, Windows 11 or Ubuntu Computer
Basic knowledge of Python, Powershell/Terminal, VS code is a must - we will use the Python SDK for ADK
Basic familiarity with LLMs like Google Gemini and your choice of OS is a must

Description

NEW ADK 2.0 is coming. Checkout the ADK 2.0 additions below!

New Update [APR 2026]: New ADK Version 2.0 - Overview of Graph Based Workflows. Note this is a new experimental feature. I am adding an overview for now and we'll learn further when ADK 2.0 actually comes into production

New Update [JAN 2026]: Added Advanced Tool Use using Programmatic tool calling with ADK + MCP!

TL;DR - Learn to build AI agents using Google’s Agent Development Kit (ADK) on Mac + Windows + Ubuntu. This course introduces you to this new agent framework, walks you through Google’s ADK, and shows you how to build, debug, and deploy intelligent agents using Python ADK. You'll get hands-on experience with real code, multi-agent systems, streaming, and tool calling.

What You Will Learn and Build

Introduction to Agent Frameworks and Agent Development Kit (ADK)
Simple agents and complex ADK Workflows like Sequential and Parallel Agents
Deployment to Google Cloud
MCP Integration (MCPToolset) along with Advanced Tool Use like Programmatic Tool Calling
Agent Config and No-Code Visual Agent Builder

Key Features of Google ADK

Software-first Agent Design: Develop agents like apps, with predictable behavior and structure
Web Playground: Use adk web to launch a full visual UI to interact with your agents
Event & Token Streaming: Real-time tracing, token-by-token output, and request inspection
Bi-directional Audio/Video: Let agents hear, speak, and see users in real-time
Model Agnostic: Use Gemini, OpenAI, or even open-source models

Important Notes

The Course is recorded on a MacBook and Windows and Ubuntu
Lectures are labeled with OS it is recorded on. When no label is provided, it means that it is a common lecture for all platforms
This course is based on Google ADK which is still evolving. The latest Python version offers stability and used for most implementations. Things might break, please be open to that. We'll address these aspects as they happen!

Disclaimer

This course is for educational purposes only. You are responsible for reviewing all terms, privacy policies, usage restrictions, and pricing for third-party services you use during this course (such as Google Gemini or GCP). We do not offer any guarantees or warranties related to these external services.

The Gemini API key and other tools are provided by Google and may change over time. We simply guide you through their use in learning environments and demonstrate practical agent development techniques.

Let’s Build the Next Generation of Agents Together!

By the end of this course, you’ll have built your own AI agents using ADK — agents that can reason, communicate, use tools, and collaborate with other agents. Whether you’re here to learn, explore, or innovate, this course is your gateway to the future of agentic computing.

Who this course is for:

Take this course if you want to learn to build sophisticated Agents with Google's new Agent Development Kit
For developers, programmers and AI Engineers looking to learn a new Agent Framework (ADK)
For entrepreneurs looking to get a quick start on ADK
For no-code developers wanting to use visual builders to build agents
For MCP developers wanting to integrate MCPToolset and Programmatic Tool Calling

Master Google ADK - Agent Development Kit (Mac, Win, Ubuntu)

What you'll learn

Explore related topics

Course content

Introduction6 lectures • 9min

Build Your First ADK Agent11 lectures • 48min

ADK Workflow (Sequential) Agent - Live Coding a Website Builder11 lectures • 1hr 3min

ADK Workflow Agents - ParallelAgent7 lectures • 35min

Deploy your Agent to Google Cloud13 lectures • 42min

Four different ways to run your ADK Agents7 lectures • 31min

Using MCP Servers with ADK Agents - McpToolset Class - Example 111 lectures • 51min

Using MCP Servers with ADK Agents - McpToolset Class - Example 222 lectures • 41min

Using MCP servers with ADK Agents - Programmatic Tool Calling - Example 37 lectures • 19min

Using No-Code to Build Agents with ADK9 lectures • 18min

Requirements

Description

Who this course is for: