Hacker News Reader

None of us have gone back to writing 100% assembler.

We would have. IF a typical C codebase in its early years sucked as much as some AI generated codebases do today.

But instead, humans preferred the productivity AND the readability AND maintainability that elevated us above assembly. Maybe AI-generated code will get there also.

As someone maintaining and using agentic AI developed codebases, they’re better maintained than human stuff I worked on.

For example, we rigidly require specs that can be used to drive end to end tests. Commits can’t break builds or tests. 100% test coverage required for both unit and integration/E2E tests.

And all of this alongside adding macOS/iOS/Android to our standard builds instead of just web/electron. Oh. And moving backends to Rust/Go.

I hear you, and you're right that rigidity and the extra work of clear specs and strong tests don't bother an LLM. When they're used. And they should be used. But experienced software developers also use those procedures and understand their value. i.e. working on a codebase that doesn't make you want to quit!

Anyways, you've obviously got good policy in place for generating code. This wisdom isn't universal.

Reply on news.ycombinator.com