Looks like the experiment was a success. We will be seeing many of these and it will work out fine.

"Code is the spec" and "tests are the success rates" - as principles for these rewrites. We can look forward to many of big complex software re-written in language that make sense.

Claude's workflows might become good training data for LLMs: https://github.com/oven-sh/bun/tree/23427db/.claude/workflow...

Maybe someday GitHub will find a way to optimize their system so I can actually see the PR and not a unicorn.

"just an experiment"