Skip to content

Vision

All systems need observability. That starts with emitting telemetry data. But the raw data alone isn’t enough. There’s a set of steps - roughly: collect, transform, interpret, act - through which telemetry data passes that are necessary for software to be not merely observable, but observed by real, actual humans. These steps are performed by an observability platform, driven by the configuration of that platform.

Of course, as software evolves, so too must the configuration that allows it to be observed. By nature, software and its observers are companions. Today’s tooling, however, makes co-evolution of the two profoundly frictionful. That’s what we’re aiming to change - no matter how you ship software, your observability can ride sidecar. Achieving this has the potential to make observability go mainstream, much as testing has over the last decade.