Cole has actually been undertaking significant work on the backend to support scaling in a way that drastically reduces costs (e.g., many server instances horizontally rather than expanding resources on singular machine). This should keep costs lower at any given level of activity.
We are also a specialized instance focused on technology and more technical subjects. This inherently helps keep costs down compared to more generalized instances that need to support incredible volumes.
He’s hoping to have more details on what he’s doing both for us users and other instances down the line. For now, he’s focused on getting through a list of further improvements and resolving any peculiarities in the Lemmy codebase that come up!