While the title says "What is a Vector Database?", the video has no content describing what a vector database actually is (e.g. how is the data stored and accessed, what embeddings are, etc.), and no details of how vector databases differ from the others (e.g. no explanation of where/when/how a vector database is more performant and/or scalable than others). Obviously, detailing these would make a much longer video, and make it a much more useful one.
A vector database is a specialized database designed to store and query high-dimensional vectors, which are often used in machine learning and data science. These vectors represent data points in a multi-dimensional space, allowing for efficient similarity searches, nearest neighbor searches, and other operations critical to applications like recommendation systems, image retrieval, and natural language processing. By organizing data in this way, vector databases can perform complex operations quickly and accurately, making them essential for handling large-scale, unstructured data. Examples include Pinecone, Milvus, and FAISS, which are tailored to manage the specific needs of high-dimensional vector data.
*Good theoretical video* except there was no sample tool, sample query or sample response in the video at all. Just like supervised learning, humans also understand fast and deep, when there are some real world examples of vector-queries and vector-response examples from one or few real world tools.
I was wondering about this as well. My guess: they flip the whole video horizontally . You'll note that there's nothing in the background that would indicate if left/right was reversed