Welcome!
1. Welcome!
Hello and welcome to this course on building AI agents with Google ADK. We're thrilled to have you here.2. Beyond a single LLM
If you’ve ever used a chatbot like Gemini to summarize an email or answer questions, you’ve already seen what a single Large Language Model, or LLM, can do.3. Beyond a single LLM
But what happens when a task is more complex, like managing customer queries,4. Beyond a single LLM
planning a trip,5. Beyond a single LLM
or troubleshooting code?6. Enter AI agents
That’s where AI agents come in. An agent goes beyond generic chat responses. It’s a purpose-built assistant that leverages an LLM to interact with its environment to achieve user-defined tasks, combining7. Enter AI agents
reasoning,8. Enter AI agents
planning,9. Enter AI agents
and the execution of actions.10. What AI agents can do
It uses LLMs under the hood, but can also call external tools or APIs, follow safety guardrails, and even collaborate with other agents to complete multi-step tasks.11. What we'll build
In this course, we’ll build one from scratch to help us manage requests from our customers. By the end of the course, our customer support agent will answer questions using a knowledge base, align its answers with internal policies, and filter out unwanted or unrelated content.12. How our agent will evolve
We’ll start by teaching our agent the basic skill of greeting users, and quickly level up to13. How our agent will evolve
building a full multi-agent system that works collaboratively to handle complex requests.14. Meet the Google ADK
We won’t be starting from a blank slate either. Google’s Agent Development Kit is a powerful open-source toolkit that gives you all the building blocks, from agent definitions to multi-agent orchestration, so you can focus on your agent’s behavior and logic, not infrastructure.15. What you'll need
This hands-on course is designed for developers who are comfortable writing Python and working with APIs. No prior AI experience is required.16. Learn by doing
We’ll walk you through the concepts and then dive into coding directly in DataCamp’s DataLab environment. DataLab is an interactive, notebook-style workspace where we can write code, see results instantly, and document our process step by step, all in one place.17. Let's practice!
We'll see our code in action right away — starting with an agent that welcomes our customers. Let’s get building.Create Your Free Account
or
By continuing, you accept our Terms of Use, our Privacy Policy and that your data is stored in the USA.