# KodeKloud: Beta Courses Google ADK

## Google ADK

### Introduction

- [Course Introduction](https://notes.kodekloud.com/docs/Google-ADK/Introduction/Course-Introduction/page.md): Hands-on course teaching Google ADK to build, deploy, and operate LLM-powered cloud automation agents with tools, structured outputs, observability, and production best practices.
- [ADK Overview](https://notes.kodekloud.com/docs/Google-ADK/Introduction/ADK-Overview/page.md): Overview of Google's open-source ADK for building production-ready multi-agent, multimodal, real-time AI applications with orchestration, observability, and deployment guidance
- [Where ADK fits in the Google cloud ecosystem](https://notes.kodekloud.com/docs/Google-ADK/Introduction/Where-ADK-fits-in-the-Google-cloud-ecosystem/page.md): Explains Google ADK as an open source, model‑agnostic framework for building production‑ready, multi-agent enterprise AI with GCP integrations, observability, and deployment tooling.
- [Build your first agent](https://notes.kodekloud.com/docs/Google-ADK/Introduction/Build-your-first-agent/page.md): Tutorial showing how to scaffold and run a minimal Google ADK Python agent with simple tools for time and weather, and how the LLM routes tool calls
- [ADK Web Interface](https://notes.kodekloud.com/docs/Google-ADK/Introduction/ADK-Web-Interface/page.md): Guide to the ADK web interface for developing, debugging, and tracing agent behavior, demonstrating a minimal agent, local server setup, invocation traces, and UI tools for performance diagnosis.

### ADK Fundamentals

- [Scaffolding our ADK agent](https://notes.kodekloud.com/docs/Google-ADK/ADK-Fundamentals/Scaffolding-our-ADK-agent/page.md): Guide to scaffold and run a minimal ADK helpdesk agent using Python, Google AI backend, and ADK tools including CLI and web UI
- [Tool Fundamentals in ADK](https://notes.kodekloud.com/docs/Google-ADK/ADK-Fundamentals/Tool-Fundamentals-in-ADK/page.md): Explains registering typed tools in ADK so LLM agents can call Python functions, return structured outputs, and integrate utilities for actionable helpdesk tasks and production best practices.
- [Implementing our Tools](https://notes.kodekloud.com/docs/Google-ADK/ADK-Fundamentals/Implementing-our-Tools/page.md): Guide to implementing and registering two typed helpdesk tools that lookup users and check service status for an ADK LLM agent.
- [Using State to Drive Conversations](https://notes.kodekloud.com/docs/Google-ADK/ADK-Fundamentals/Using-State-to-Drive-Conversations/page.md): Guide on using ADK session state to manage multi-turn conversational agents, tracking minimal curated values, scoping keys, dynamic prompt injection, and deterministic state updates for troubleshooting flows.
- [Adding State to the IT Helpdesk Assistant](https://notes.kodekloud.com/docs/Google-ADK/ADK-Fundamentals/Adding-State-to-the-IT-Helpdesk-Assistant/page.md): Explains building a stateful IT helpdesk assistant that uses session history and targeted tools to clarify issues, check service and account status, and provide concise troubleshooting steps.

### Deploying and Operating ADK Agents

- [Why Structured Outputs Matter and How ADK Handles Them](https://notes.kodekloud.com/docs/Google-ADK/Deploying-and-Operating-ADK-Agents/Why-Structured-Outputs-Matter-and-How-ADK-Handles-Them/page.md): Explains why structured outputs are important and how ADK uses agent schemas and tool-level structured results for reliable help desk ticketing workflows.
- [Implementing Ticket Schemas](https://notes.kodekloud.com/docs/Google-ADK/Deploying-and-Operating-ADK-Agents/Implementing-Ticket-Schemas/page.md): Refactoring an agent demo to use Pydantic ticket schemas and typed tools so the agent orchestrates while tools handle validation, lookups, and ticket creation.
- [Error Handling Resilience](https://notes.kodekloud.com/docs/Google-ADK/Deploying-and-Operating-ADK-Agents/Error-Handling-Resilience/page.md): Guidelines for implementing structured error handling in LLM agents and tools to produce predictable, machine readable failures and helpful user facing messages while keeping sessions resilient
- [Evaluations](https://notes.kodekloud.com/docs/Google-ADK/Deploying-and-Operating-ADK-Agents/Evaluations/page.md): Describes a lightweight, event-driven evaluation workflow for ADK agents measuring tool trajectory and response similarity using captured golden sessions via UI or CLI