Why build Event-Driven AI systems?
AI Summary
In this video titled “Why build Event-Driven AI systems?”, Adi Polak of Confluent Developer explains the architecture and practical implementation of AI agents using event-driven microservices. The presentation covers key concepts including agentic architecture, predictive sales through a case study of Badflix, the roles of different AI system components, handling feedback loops, and integrating real-time data for customer predictions. Polak discusses the transition from monoliths to scalable microservices, the benefits of using Kafka in an event-driven architecture, and the integration of Flink for streaming data. The video also explores the governance, schema registry, and observability necessary for managing AI systems effectively. Additional resources and links are provided in the video description.