Close. It means there's no persistent infra charges and you're charged on use. You dont run anything permanently.

It still doesn't capture the concept because, say, both AWS Lambda and EC2 can be run just for 5 minutes and only one of them is called serverless.

Unless the engineer takes steps to spin down EC2 infrastructure after execution, it is absolutely persistent compute that you're billed for whether you are doing actual processing or not. Whereas lambda and other services are billed only for execution time.

Drywall is pretty amazing, but I don't agree with all the points in the article.

It cheap to buy and cheap to install, easy to cut and installs fast. It's tolerant with imperfect walls and is surprisingly flexible. It can also be seamlessly repaired.

It can also act as a primary air barrier.

I do not like moisture resistant drywall, moisture control is more important as well as using proper materials in high humidity areas.

Because drywall is cheap, incredibly tolerant of movement and irregularities. It's also super easy to repair. It can also act as an air barrier for energy efficiency. A drop ceiling is terrible for that and is ugly AND expensive.

It's more important to control moisture than buy moisture resistant materials.

Drywall is great, its cheap and easy to repair.

I would not want to live in any water damaged house without remediation due to the risk of mold.

Everything costs money. It's not like drywall doesn't allow you to soundproof, 2x layers of 5/8" is a common method as well as staggering 2x4's in a 2x6 wall.

Wood isn't flimsy. Drywall is fine, it's cheap and easy to repair. If built correctly can last 100+ years.

Sure, and you may as well walk around with a blindfold on to develop your "spidey" senses too.

Humans can't handle large tasks either, which is why you break them into manageable chunks.

Just ask claude to write a plan and review/edit it yourself. Add success criteria/tests for better results.

Making clones of production isn't trivial. Is your app server clone going to connect to your production database? It is going to spin up your whole stack? Seems a bit naive.

A better approach is to have AI understand how prod is built and make the changes there instead of having AI inspect it and figure out how to apply one off changes.

Models are already very good at writing IaaC.

FYI, this is based on libSQL which has pretty poor driver support (Python still experimental).