​(Don't) Play It Again: Recommendation Engines from Data to Prediction

​One evening, one dataset, one idea: if the algorithm knows what you've listened to, can it guess what you want to hear next?

Play It Again is a hands-on workshop where we build a music recommendation engine from scratch. We'll start with raw Spotify track data, clean and shape it together, then wire up a model that actually makes predictions. You follow along, you build, you leave with something that runs.

You'll be in the code the whole time, working with real data on a problem worth solving. Expect to build.

​What we'll build

​♫ Data prep: Pulling apart the Spotify dataset, understanding what's in it and getting it into a shape a model can use.

​🤖 Recommendation engine: Training a model on listening patterns and turning it into something that suggests what to play next.

📊 Results you can see: Exploring the dataset visually with charts so you understand what you're working with. Then seeing the model's actual output as a ranked list of track recommendations you can read and reason about.

​How the night runs

• ​Doors open at 6PM.

• ​Kickoff at 6:30 PM with a short intro, then we get into it.

• ​Two parts, one flow. Data first, predictions second.

• ​Food and drinks included.

​Bring a laptop. We'll work in Google Colab, so no setup needed ahead of time.

Come to learn, come to build, come to see your playlist in a completely different light.

​By taking part in this event, you grant the event organizers full rights to the use of images resulting from the photography/video filming, and any reproductions or adaptations of the images for fundraising, publicity or other purposes. This might include (but is not limited to): the right to use them in printed and online publicity, social media and press releases.

​If you do not wish to be photographed, please inform an event organizer during check-in.