Designing Data Layers for Agentic AI:
Patterns for State, Memory, and Coordination at Scale

Discover the Power of a Unified Data Layer for Agentic Applications

AI agents introduce a new class of architectural challenges: managing shared memory, coordinating state across agents, and ensuring auditability in distributed, long-running workflows.

Why Watch?

In this session, we explore emerging data layer patterns for agentic systems, including approaches to conversation state, knowledge persistence, and cross-agent coordination. We’ll examine the tradeoffs between event-driven and stateful designs, and how consistency, latency, and cost shape system behavior at scale.

What You'll Learn

Using a reference architecture built on AWS and YugabyteDB, we’ll show how these patterns can be implemented in practice, highlighting design decisions, failure modes, and operational considerations for production-grade AI systems.