
Real-time retrieval layer for conversational AI
moss.dev (opens in a new tab)Moss is a retrieval layer for conversational AI that hands a product the right context at the right moment without slowing it down, so the experience feels instant. It runs in the browser, at the edge, on the device and in the cloud. Moss is deployed in roughly 150 countries and has served more than 10 million minutes of real-time voice, and a growing share of its customers are moving from evaluating it to running it in production.