Building change-driven solutions that react to specific events across distributed data systems can be challenging. Engineers often end up stitching together polling loops, custom listeners, and brittle glue code just to keep downstream systems in sync. This talk introduces Drasi, a CNCF Sandbox project that codifies continuous query and reaction patterns, making it dramatically easier to design and implement architectures that respond instantly to the right data changes—without the operational overhead.
We’ll explore how Drasi enables developers to declaratively define what data changes matter and how their systems should react, allowing them to focus on business logic rather than infrastructure plumbing. Beyond day-to-day operational scenarios, like keeping microservices and operational dashboards consistent, Drasi also serves emerging AI patterns. As models increasingly depend on constantly evolving datasets, ensuring that the data powering AI pipelines, retrieval systems, and real-time decision engines stays up-to-date becomes critical. Drasi provides a consistent, scalable way to detect relevant changes and trigger model refreshes, embeddings updates,and more.
Attendees will walk away with a clear understanding of change-driven architecture patterns, how Drasi implements them, and how to apply Drasi to both practical distributed-system problems and modern AI-driven workflows. Live demos will show Drasi in action, from simple event-driven reactions to keeping data fresh for AI applications.
Audience:
Developers, architects, SREs, and AI practitioners interested in change-driven systems, distributed data, and practical patterns for maintaining up-to-date state in modern applications.
Takeaways:
- Understand the core principles of change-driven architecture
- Learn how Drasi codifies continuous queries in a declarative manner
- See real-world examples across operational and AI workloads
- Gain practical guidance for applying Drasi to your own systems
