Your distributed systems hum along happily, everything is consistent. Until they aren’t. One tiny change later and… UH OH. Logs explode into chaos, dashboards blink like disco lights, and you’re neck-deep in debugging an issue that seems to make no sense. The data looks wrong, the system disagrees with itself, and you’re stuck wondering: how did this even happen?

We learnt this the hard way while building Kratix, a platform framework that sits on top of Kubernetes.

When Sid, one of our favourite users at a bank, did a demo for his org and showed the number steps he took to find the root cause of a problem in our software,, we were ashamed at ourselves for not doing anything sooner.

Our customers build complex, distributed platforms, and for a long time felt the same pains as Sid while using our product. When they kept coming running to us asking for solutions, we knew we had to do something about it - because learning how to observe eventually consistent, distributed systems is HARD. Correctness is a journey, not a guarantee.

In this talk, we will walk you through proven strategies for making eventually consistent systems less mysterious and more observable, and how these strategies were applied in the Kratix framework. You’ll learn how to:
- Add meaningful, traceable errors that actually help humans
- Turn events and state transitions into your best debugging tools
- Build dashboards that explain system health, not just CPU usage
- Set up alerts that reflect business impact, not just technical noise

By the end, you’ll be ready to turn your system’s logs, metrics, and states from a source of confusion into a source of clarity, no matter what part of the stack you work on.