AR
arjunprabhulal/google-adk-masterclass
Build AI agents with Google's Agent Development Kit (ADK). 20 modules from basics to advanced topics.
🌟 Google ADK Masterclass
📖 Overview
Build production-ready AI agents with Google's Agent Development Kit (ADK). This hands-on masterclass covers everything from basic setup to advanced topics like multi-agent systems, memory management, and custom integrations.
What this project offers:
- 📚 Step-by-step tutorials progressing from fundamentals to advanced techniques
- 💻 Modular, ready-to-run code examples with minimal setup
- 🏗️ Reference architectures for building real-world AI agent applications
▶️ Demo
💡 Why This Exists
Learn ADK through hands-on projects. Each module pairs concepts with working code, progressing from basics to advanced multi-agent patterns.
🎯 Who Is This For
| If you are... | This is for you |
|---|---|
| 🆕 New to AI agents | Start from Module 1, learn foundations first |
| 🔄 Coming from LangChain/CrewAI | Jump to Module 4+ for ADK-specific patterns |
| 🏗️ Building production systems | Focus on Modules 16-20 for state, memory, events |
| 🔧 Integrating tools/APIs | Modules 7-14 cover built-in and custom tooling |
✨ Features
- 📋 Comprehensive curriculum — 20 modules covering agents, workflows, tools, memory, and integrations
- 🖥️ Multiple interfaces — CLI, Web UI, and programmatic approaches for every example
- ⚙️ Production patterns — state management, callbacks, artifacts, and event streaming
- 🔧 Extensible tooling — built-in tools, custom functions, OpenAPI, and MCP integrations
- 📜 MIT licensed — free to use, modify, and deploy for personal or commercial projects
📚 Modules
🏁 Foundations
| # | Topic | Description | Blog |
|---|---|---|---|
| 01 | Getting Started | Introduction to ADK, environment setup, first agent | Read |
| 02 | Setting Up Agents | CLI, Web, and Programmatic setup methods | Read |
| 03 | Visual Builder | No-code agent building with Visual Builder | Read |
🤖 Agent Types
| # | Topic | Description | Blog |
|---|---|---|---|
| 04 | LLM Agents | Building intelligent LLM-powered agents | Read |
| 05 | Workflow Agents | Sequential, Parallel, and Loop patterns | Read |
| 06 | Multi-Agent Systems | Agent orchestration and collaboration | Read |
🔧 Tools & Integrations
| # | Topic | Description | Blog |
|---|---|---|---|
| 07 | Built-in Tools | Google Search, Code Executor | Read |
| 08 | Vertex AI RAG | RAG Engine integration | Read |
| 09 | Vertex AI Search | Enterprise search integration | Read |
| 10 | Custom Function Tools | Building custom Python tools | Read |
| 11 | OpenAPI Tools | REST API integration | Read |
| 12 | Multi-Tool Agent | Combining multiple tools | Read |
| 13 | Third-Party MCP Tools | GitHub, Firecrawl integration | Read |
| 14 | MCP Toolbox for Databases | Database integration with MCP Toolbox | Read |
Protocols
| # | Topic | Description | Blog |
|---|---|---|---|
| 15 | Model Context Protocol | MCP architecture and patterns | Read |
⚙️ Core Components
| # | Topic | Description | Blog |
|---|---|---|---|
| 16 | Session, State & Memory | Conversation history and state management | Read |
| 17 | Context Management | Caching and compaction | Read |
| 18 | Callbacks | Intercepting agent behavior | Read |
| 19 | Artifacts | File and data handling | Read |
| 20 | Events | Event streaming and debugging | Read |
📋 Prerequisites
For most modules:
- Python 3.10+
- Gemini API key from AI Studio
For Vertex AI modules (08, 09, 18):
- See GCP Setup Guide for detailed instructions
🚀 Quick Start
# Clone the repository
git clone https://github.com/arjunprabhulal/google-adk-masterclass.git
cd google-adk-masterclass
# Set up environment
echo "GOOGLE_API_KEY=your-api-key-here" > .envOption 1: Using uv (Recommended)
# Install uv if you haven't already
curl -LsSf https://astral.sh/uv/install.sh | sh
# Create virtual environment and install dependencies
uv venv && source .venv/bin/activate
uv pip install -r requirements.txt
# Run your first agent
cd 01-getting-started
adk webOption 2: Using pip
# Create virtual environment
python -m venv .venv
source .venv/bin/activate # On Windows: .venv\Scripts\activate
# Install dependencies
pip install -r requirements.txt
# Run your first agent
cd 01-getting-started
adk web🔗 Resources
🤝 Contributing
Contributions are welcome! Whether it's:
- 🐛 Bug fixes
- 📝 Documentation improvements
- 💡 New module suggestions
- 🌟 Sharing your projects built with this masterclass
Please open an issue first to discuss what you'd like to change.
👤 Author
Arjun Prabhulal
- 🌐 Website: arjunprabhulal.com
- 💻 GitHub: @arjunprabhulal
- 🐦 Twitter: @arjunprabhulal
📄 License
This project is licensed under the MIT License - see the LICENSE file for details.
⭐ If you find this helpful, please star this repository and share with others!