AI agents are revolutionizing the analytics landscape, enabling NLP-driven conversations and transforming user interactions with data. However, our traditional engineering priorities and architectures must evolve to support this paradigm shift. In this session, we will explore the critical architectural changes required to empower AI agents, focusing on Retrieval-Augmented Generation (RAG), metadata richness, the AI semantic layer, and vector databases. We’ll discuss the role of knowledge graphs, feature stores, and Unified Observability concepts. Finally, we’ll map out an AI-agentic architecture, integrating large language models (LLMs), event-driven data pipelines, and fine-tuned embeddings, drawing parallels with existing technologies to pave the way for the next era of data-driven intelligence