This sparks a thought experiment I've been having. In this world where llms can be thought of as the new layer of compilers, things like pickle are likely going to be the main unit of work for humans.

Only now instead of this developing brittle generated tests, it will instead be used by the llm as guidance to generate the actual code and tests.

Before people jump down my throat, I know we are nowhere near that today and I promise I'm not pitching this to my leadership because they would gobble it up too fast.

But for us engineers, I think there is an interesting space for thinking of llms as akin to garbage collection, a feature that allows us to abstract to a slightly higher level of thought. Yes we still need to know how to check under the hood, but this is looking like the right level of precision-flexibility ratio that llms thrive in

I'm building this over at https://zapcircle.com/ - it's still a work in progress, but it's all open source.

The idea behind it was that Behavior-Driven Development might be a great idea, but Gherkin was a pain to work with. LLMs bridge that gap now:

https://www.jefflinwood.com/2025/zapcircle-bdd-2025/