Vector databases are all the rage today. I’ve built a few iterations of vector search, beginning in 2011 at Artsy, powered by the Art Genome Project. The first attempt was a brute-force k-nn with data stored in MongoDB, written in Ruby, which was improved used LSH, then NN-Descent. Around 2017 we migrated to Elasticsearch that had some specialized k-nn support. Things have evolved rapidly with generative AI, so let’s try to index and search some vectors in 2023, in Python, and without using any ...