Operationalizing Intelligence: Turning Event Streams into Event World Models
Horst Samulowitz, IBM Research
Thu, Nov 5, 9:30–10:15 AMRoom 2012
Organizations use event streams to monitor systems, trigger workflows, and power real-time applications. But event streams also contain the information needed to answer harder questions: What is the current state of the system? Where is it heading? What is likely to happen next? How would the outcome change if we intervened?
This talk introduces Event World Models, an approach that transforms streams of business activity into a continuously evolving representation of enterprise state and behavior. By organizing records into entities, actions, relationships, and processes, Event World Models enable applications to understand how a system evolves over time and reason about the consequences of potential actions.
We’ll discuss key architectural patterns, show how these models can be maintained directly from event streams, and walk through practical examples. Attendees will learn how Event World Models extend traditional event processing and open the door to a new generation of intelligent streaming applications.
Speaker
-
Horst Samulowitz
Principal Research Scientist and Manager, IBM Research
Horst Samulowitz is a Principal Research Scientist and Manager at IBM Research, where he leads work on generative AI for enterprise data systems. He previously conceived and led IBM’s AutoAI and Semantic Automation initiatives and is currently focused on autonomous data systems that combine large language models, agents, and world models to discover, understand, and manage enterprise data. Horst holds a Ph.D. in Computer Science from the University of Toronto and has held research positions at Microsoft Research and the University of Melbourne.