miniblog.

I find that I'm choosing AI tools based on the quality of the harness rather than the model. For example, I'm using Claude Code Web because its model of remote VMs is extremely convenient. I'm using Perplexity because it's tuned really well for web searches.

LLVM's lit tool for testing files is the most successful testing DSL I've ever seen. It's used a decent number of projects outside of LLVM. By contrast I don't see many folks using e.g. Cucumber these days.

Anthropic used Claude to write a C compiler in Rust (CCC). I've really enjoyed following the discussion. The initial post shows how they wrote the harness with GCC as an oracle and a prompt specifying a design using SSA. The author has clear domain expertise. https://www.anthropic.com/engineering/building-c-compiler