From RAG to Relational: How Agentic Patterns Are Reshaping Data Architecture

In this episode of the Data Engineering Podcast, Tobias Macey interviews Marc Brooker, VP and Distinguished Engineer at AWS, about how agentic workflows are impacting database usage patterns and architectural requirements. They discuss the role of databases in typical agentic workflows, the changing requirements for database systems due to agents as builders and operators, and how the choice of persistence layers is affected by agent autonomy and model capabilities. Brooker also touches on the benefits and tradeoffs of using serverless functions like Lambda for LLM calls, the geo-distributed nature of dSQL, and the shift from local to remote MCP endpoints. He shares insights on the unpredictable nature of model capabilities, the importance of data security and privacy, and the increasing role of object storage in AI applications.