Google Developers Codelabs provide a guided, tutorial, hands-on coding experience. Most codelabs will step you through the process of building a small application, or adding a new feature to an existing application. They cover a wide range of topics such as Android Wear, Google Compute Engine, ARCore, and Google APIs on iOS.

Codelab tools on GitHub

Filter By

Choose a topic
Choose a product
Choose an event

Updated August 28, 2025

In this codelab we will deploy several agent services using cloud run and agent engine,then inspect how we implement A2A protocol to standardize communications between them. You will learn the core concepts and components of A2A and inspect the interactions by deploying the agents powered by different agentic framework with Gemini backend as different services on top of Cloud Run and Agent Engine

Updated August 28, 2025

Learn how to build and deploy a secure Model Context Protocol (MCP) server on Cloud Run. Then authenticate to the remote MCP server and invoke MCP tools and prompts from Gemini CLI.

Updated August 28, 2025

Learn how to connect your deployed ADK agent to a deployed Gemma model for scalable AI applications.

Updated August 28, 2025

As the DevOps/SRE Guardian in the four-part Agentverse series, you will build a complete, production-grade AgentOps bastion for deploying and managing AI agents. You will deploy GPU-accelerated LLMs, secure them with a Model Armor gateway, automate their lifecycle with a CI/CD pipeline, and establish deep observability.

Updated August 28, 2025

As the Data Engineer workshop in the four-part Agentverse series, you will transform chaotic, unstructured text into a powerful, structured knowledge engine. You will then build and deploy a wise Retrieval-Augmented Generation (RAG) agent that can provide nuanced, context-aware answers from this new wellspring of data.

Updated August 28, 2025

As the Architect in the four-part Agentverse series, you will master multi-agent orchestration by using the Google Agent Development Kit (ADK) to design and deploy complex, stateful workflows, creating specialist agents. And using the Agent-to-Agent (A2A) protocol to discover and delegate complex tasks to your new agent legion.

Updated August 28, 2025

As the Developer workshop in the four-part Agentverse series, you will master “vibecoding,” using the Gemini CLI to translate a high-level design document into a fully functional AI agent. You will then forge a complete CI pipeline with the Agent Development Kit to automatically test, containerize, and prepare your battle-ready agent for deployment.

35 minutes

Updated August 28, 2025

This hands-on codelab is your first step into building AI agents with Google’s Agent Development Kit (ADK). You will learn how to set up your development environment on Google Cloud, define the core logic for a basic conversational agent powered by a Gemini model, and configure its settings. By the end, you’ll have built and tested your very own “personal assistant” agent, interacting with it through both a command-line interface and a user-friendly web UI, creating a solid foundation for more advanced projects.

49 minutes

Updated August 27, 2025

One of the most common features of a website is displaying a Google map that highlights one or more locations for a business, establishment, or some other entity with a physical presence. How these maps are implemented can vary greatly depending on

59 minutes

Updated August 27, 2025

Imagine you have many places to put on a map and you want users to be able to see where these places are and identify which place they want to visit. Common examples of this include: In this codelab, you will create a locator that draws from a live