Vector databases have become very popular in the AI era. Particularly the feature of finding semantically similar objects is quite useful in AI and related applications. But how does a vector database do it? How exactly does a vector database store its data so that it is able to perform a semantic search so fast?

This talk will introduce vector databases while explaining the fundamental concepts behind it. It will explain what vectors are, how are they stored, what are vectorization techniques, indexing algorithms and similarity/distance metrics.

In order to explain these in simple terms, we start with simple use cases that are implemented with vector databases and then slowly increase the complexity to take it to mind-blowing level!