In this workshop, we want to explore what AI agents are, how they are technically structured, and what you can do with them. To do this, we will first look at some architectural patterns.

Once the theory is clear, we will put together our own agent and try it out by triggering actions and letting the AI make decisions on its own. No programming knowledge or special software is required, but you will need a laptop. It will be very hands-on!

Target Audience: anyone interested in the topic
Prerequisites: Laptop with internet access (smartphones are not suitable), browser
Level: Advanced

Extended Abstract:

At the beginning of 2024, ChatGPT was not yet able to provide information about today's weather in Paris. The chatbot only accessed the training data, and the cutoff was months ago. Today, we only have to wait a moment, and ChatGPT quickly searches for it on the internet. How is this possible?

The explanation is simple: ChatGPT is no longer just a chatbot, but an AI agent. AI agents are a hot topic in research and development and will have a significant impact on our working environment.

In this interactive workshop, we will look at what AI agents are, how we can use them effectively, and what advantages this use brings. To make things more tangible, we will assemble and try out our own agent in a no-code environment. If everything works as expected, the agent will search a file, use a calculator, and call various API functions.

At the end, participants will have seen both the strengths and weaknesses of AI agents in SW testing with their own eyes.