This week's episode is in English. That is because we had Kamilla Kurta in the studio. She is Machine Learning Customer Engineer at Google and came by to tell us everything we wanted to know about Google Agent Development Kit (ADK). As the name already indicates this is an environment in which developers can build, deploy and monitor AI agents.

At the end of October (on 26 and 27 October, to be exact), Google hosts an Agents Retreat in Rotterdam. There, developers can learn all there is to learn about Google ADK. This is a recent addition to the porftolio of Google, and was launched at the most recent edition of Google Cloud Next, earlier this year. 

Google ADK is open-source, and has been donated by Google to the Linux Foundation. However, Google is still actively involved in it. The event later this month in Rotterdam is testament to that. In this episode of Techzine Talks, we discuss various components of Google ADK, and try to go a bit more in-depth into some of the concepts.

Among other things, we discuss the following topics:

* What is Google ADK? What makes it different when it comes to the development of AI agents?
* Foundational concepts of building AI agents
* How do you deploy and evaluate agents?
* AI agents need memory too. What are the different types of agentic memory and what are the challenges they bring?

 "Calling all developers! Are you ready to take a deep dive into learning the fundamental concepts, designing and deploying truly intelligent and problem solving AI Agents in your organisation? Then join the Agents Retreat and learn how to build your own autonomous solutions!"