Searching Vector Embeddings at Scale with Weaviate

Learn about efficient vector embedding search and storage in this 25-minute tech talk from Weaviate's Dan Dascalescu. Explore how the rise of Large Language Models (LLMs) like GPT, LLaMA, Chinchilla, and Claude has created a growing need for robust vector embedding storage and indexing solutions. Discover how Approximate Nearest Neighbor (ANN) algorithms enable efficient searching of top matching results, and understand the challenges of maintaining separate databases for objects and their embeddings. Dive into Weaviate's open-source vector database solution that uniquely combines object and vector storage, enabling filtered vector searches and streamlined database operations. Master the implementation of question answering, multi-modal search across text, image, and audio, and learn to integrate third-party APIs like OpenAI for generative search capabilities. Gain practical knowledge about vectorizing data using various services including OpenAI, Cohere, or Hugging Face transformers, with the flexibility to incorporate custom vectors.