Nate Stewart

How do you keep an agent's understanding of the world fresh enough for it to act on? Interactive agents need current context about the customers, orders, inventory, accounts, and other entities they reason about. But assembling that context from constantly changing operational systems is complex, forcing teams to choose between stale data, expensive recomputation, or complex pipelines. This talk will show how to build continuously updated context in SQL that is trustworthy enough for agents to rely upon. You'll learn how to transform operational data into agent-ready representations, keep attributes and relationships up to date as the underlying data changes, and propagate those changes to search indexes and embeddings when needed. Attend if you're deploying AI agents to production and want to give them fresher, more complete context and tighten their read-act-react loop. By attending this session, your contact information may be shared with the sponsor for relevant follow up for this event only.

Modern AI applications require more than historical data - they need live, continuously updated context. This session examines how enterprises are enabling AI systems with fresh, trustworthy data to improve accuracy, relevance, and reliability. Using a real-world example inspired by production deployments, we’ll show how teams build scalable, secure context pipelines that evolve with changing data and business needs.