This repository contains code for an example content recommendation system using vector search in Tinybird.
Vector search is a great way to approach content matching and recommendations. You can calculate embeddings based on multi-modal analysis of text, images, and other media, then calculate vector distances between embeddings to recommend matching content.
Read more about vector search use cases here.
This repository mirrors work that Tinybird has done to recommend related blog posts on the official Tinybird Blog.
Read the guide to deploy this demo application yourself.
This code is available under the MIT license. See the LICENSE file for more details.