In this talk we will dive into the question whether generative AI can generate working code when we only provide test cases. So we take the main principle from TDD: we write the tests up front. Then we use a generative AI to create the code. If you wrote the tests correctly and they all pass, you should have a working prototype. Even better, if the tests pass and you trust the tests, the code is production ready!

With some live coding, we will explore this idea together. Furthermore we will take a look at the implications of this paradigm shift. After attending you should be able to apply this technique to your own project.

The title says it all! In this talk, we will explore the gaming capabilities of Large Language Models (LLMs). Since the release of ChatGPT, these models have seen steady improvements. Given that the second 'L' in LLM stands for 'Language,' it’s fitting that one of the most obvious games to play with an LLM is a text adventure. By the end of this talk, you’ll discover whether, and how well, a LLMs can navigate and play text adventures.