Generative AI is cool. Sure. But it is also limited. LLMs' power shines through if you combine their knowledge with your Generative AI, which is impressive. Astonishing. Almost magical. But, much like a GPS that insists you drive into a lake, it has limitations. The real magic happens when you combine AI's knowledge with your systems, giving it just enough guidance to be helpful without it going all HAL 9000 on you.

Enter Semantic Kernel, a way to make AI sound smart and actually be smart by calling on the world for answers when it gets confused. Think of it as a polite Hitchhiker's Guide to the Galaxy, but for your data.

So, how does this help you? Simple. AI is going to help us generate the ultimate Spotify playlist, one that perfectly matches our mood, whether we’re pounding the pavement like Springsteen or just lounging on the couch contemplating the meaning of life (and whether 42 really is the answer).

This session is mostly live coding, which means we’ll be building this system on stage, in real-time, with all the suspense of a Monty Python sketch, except with fewer Spanish Inquisitions and more AI magic. So, whether you came here to learn about AI or just want an excuse to listen to better music, this is your stop. existing systems.

Semantic Kernel is a great way to make that happen. But what does all this mumbo-jumbo mean?

In essence, it is simple: AI does what it can do, and as soon as it gets confused, it calls on the world for answers.

That gives us a nice hook to achieve what we want to achieve: generate a Spotify Playlist that matches our mood, which we can listen to while we run or while we relax on the couch, whatever tickles your fancy.

This session is mostly code, where we will build most of this system live on stage. So that's already a great way to get your heartbeat in higher gear!

Can of course also be done in Dutch,