Watch Lino’s session at the Azure Cosmos DB Conf 2026. It just passed 25,000+ views on YouTube.
Lino explains the distinction between short-term and long-term memory in LLM applications, demonstrating how to implement chat history using the Microsoft Agent Framework and Azure Cosmos DB. The session further explores techniques for improving efficiency and reducing costs by integrating semantic caching, allowing for faster AI responses through vector similarity searches.
