This demo built by VMware Data, uses the open source pgvector extension of Postgres in the VMware Greenplum Data Warehouse to show how unstructured data, such as images and text, can be used to build powerful search engines.
Dataset source: https://www.kaggle.com/datasets/akshaypawar7/millions-of-movies?rvi=1