Abstract
Real-time has come to Multi-modal AI with Google Gemini’s Flash 2.0 API. This talk will present a simple Java example of real-time AI on video and audio. You will learn why the best patterns for reactive, real-time AI are Event Sourced architectures, as LLMs are inherently event driven. In addition, you will learn how you can quickly build a Java application in Akka for real-time multi-modal Generative AI with Retrieval Augmented Generation, using the power of the Akka platform to build and deploy in a highly resilient, elastic and multi-region runtime architecture.

Benefits:
Understand challenges and best practices – for distributed Generative AI applications that are

• Event-driven, elastic and performant
• Resilient up to 6 Nines
• Ultra-low p99 latency
• Agile and flexible
• Multi-Cloud

Learn how event sourcing and Akka make these easy

Speaker
John DesJardins is Field CTO at Akka, responsible for advising customers on solution architecture and best practices. He was formerly CTO at Hazelcast, setting technology strategy for the real-time data platform. His expertise in large scale computing spans Big Data, IoT, ML and cloud. John brings 25+ years of experience in architecting and implementing global scale always-on computing solutions with Global 2000 while at Hazelcast, Cloudera, Software AG and webMethods. He is an active writer and speaker, and holds a BS in Economics from George Mason University, where he first built predictive models, long before that was considered cool.