Right. The blog post is a good narrative over time but isn't super clear about how they solved the problem. If I understand correctly they broke down the big MySQL query into separate queries that the Go service processes/caches?
Blogpost author here. That is correct. Sorry if the explanation isn't super clear; for this particular question, you can consider the two diagrams as a before and after. They pretty much convey what you have explained here.
So if the alternative was Scala I can see why Go may have helped tighten things up a bit.