Jon Bratseth on Vespa AI: Reinventing Search for Machines with RAG at Enterprise Scale

In this episode, we welcome Jon Bratseth, CEO of Vespa.ai, and Ragnar Harper, Head of Technology at AWS Norway. Jon shares Vespa's innovative approach to enterprise-scale AI, focusing on Retrieval Augmented Generation (RAG) to enable models to access up-to-date, company-specific data. He explains how Vespa's platform combines textual search, vectors, and machine learning signals to achieve accuracy and scalability, while also handling the complexity of multimodal retrieval. Jon discusses how Vespa’s open-source nature and partnership with AWS leverage advanced hardware like Graviton 4 for efficient scaling. The conversation also covers Vespa's leadership in secure data handling, particularly for enterprises dealing with sensitive information. Jon shares insights from his leadership journey, transitioning from Yahoo to startup CTO, and emphasizes the crucial role of delegation. Finally, they explore the future of Vespa, including two-layer ranking to optimize content chunk retrieval and the impact of RAG on AI-driven decision-making.