multi-modal apps via deepmind AI studio are VERY under-hyped with just a single prompt i built this app in <10 min that recorded me flipping through records and outputs every artist + album shown. video used to be one of the hardest things to work with, now it’s a prompt.

for those wondering, this is the prompt i used: "create an app that takes a video of a person flipping through their record collection and extracts the album and artist names of every album shown. you could do this by taking the video and first extracting the frames that show distinct vinyl records, and then have the vision model analyze those frames to extract the information"