When scaling up looks like scaling down

July 15, 2021

I was listening to a podcast episode where the guest was explaining problems scaling to hundreds of engineers, and one of the tools they found useful was essentially automatic linters for configuration files (Kubernetes manifests in this case).

This got me thinking: Sometimes the same tools that help an organization scale to hundreds or thousands of engineers are the exact same tools that help scale down, perhaps all the way to one engineer.

The main problem I gather the podcast guest was aiming to solve for large teams was one of communication. Teaching hundreds of developers to all follow the same standards for correctness is a daunting task. But by solving that problem, they also solved another problem that applies equally to small teams: Providing shorter feedback loops, and automating toil.