Pub/sub is where most teams begin their asynchronous journey—and often where they stop. While it’s great for decoupling services, pub/sub alone doesn’t protect you from lost messages, broken workflows, or systems that collapse under load.

In this talk, we’ll go beyond simple event broadcasting and look at the messaging patterns that production systems actually rely on. We’ll explore how the Outbox pattern prevents data loss during crashes, how sagas coordinate long-running workflows without distributed transactions, how scatter-gather unlocks parallelism for faster results, and how competing consumers keep throughput high during traffic spikes. Along the way, we’ll examine routing strategies and operational safety nets like retries, dead-letter queues, and back pressure.

You’ll leave with practical guidance on choosing the right messaging patterns to move from “event-driven” to truly resilient, fault-tolerant distributed systems.