Pinecone is a fully managed, horizontally scaling database for working with vector embeddings in production. It generates, indexes, and retrieves vector embeddings (using similarity search) within milliseconds. We are engineers who built large machine learning platforms, databases, and search engines at AWS (SageMaker), Facebook, Yahoo, and Google. We are scientists who transformed businesses with ML applications such as shopping recommendations, online advertising, semantic search, anomaly detection, and more. We are researchers who collectively published more than 100 academic papers and patents on machine learning, systems, and algorithms.