After months of testing and iteration, I have finally nailed this down. Sharing my notes before I forget the details.
Background
Before I get into the results, you need to understand the starting point. Every case study that skips context is useless because you cannot assess whether the results apply to your situation.
The messy middle
Getting started is easy. Getting to production-ready is hard. The difference is in the details that nobody blogs about because they are boring. But boring is what keeps systems running at 3 AM.
Results
Numbers. Real, unembellished numbers. I am including the failures alongside the successes because cherry-picking wins is dishonest and unhelpful.
What I would do differently
Hindsight is 20/20, but some lessons only become clear after the fact. If I were starting from scratch today, here is what I would change.
I will update this article as things change — and in this space, things change fast. Bookmark it and check back. More related content at clawgo.net, agent101.net. Official references: Let’s Encrypt, Node.js Docs.