2 posts tagged with "reliability"

A lot of AI-generated code fails the same way: it looks reasonable, maybe passes a few checks, and then falls apart the moment the real system touches it. The easy explanation is "the models aren't good enough yet." The more useful one is that most languages still assume a human is carrying the missing context in their head.

Hand an agent five equivalent patterns, prose-only errors, implicit side effects, and a flaky test surface, and it has to improvise at exactly the points where you need it to be mechanical.

That is why the recent argument for agent-first languages is worth taking seriously. Armin Ronacher's essay A Language For Agents made the thesis explicit; my own take is a bit more narrow. The reliability gap shows up wherever the language and toolchain leave too much ambiguity at the repair boundary.

Series navigation: Post 1 of 3 · Next: How X07 Was Designed for 100% Agentic Coding

For the last twenty years, most programming languages and most software practices were designed around a simple assumption: a human is the one holding the whole thing together.

A human reads code, remembers conventions, notices weirdness, and makes judgment calls when the codebase offers five equally valid ways to solve the same problem.

A coding agent works differently.

It is very good at wide edits. It is very good at following explicit contracts. It is very good at retry loops. But it is much worse than a strong engineer at carrying a large unstated architecture around in its head.

So working with an agent is not human programming sped up. It is a different optimization problem, and the things you should constrain are different too.