💭 I find that I script°hard word out an entire°hard word conversation, sometimes returning the interesting parts before the other person is ready to receive them.
💡 that is: she says her entire°hard word side of the conversation before you get a chance to talk.
🔥 this is how i do my emails.
----
⚙️ Claude°hard word is writing the README°hard word files°hard word for the React°hard word and Swift apps. This means we are about to launch°hard word.
----
💭 My findings have been, that a policy°hard word of only LLM-written°hard word code°hard word is possible today.
✨ It is the equivalent°hard word of a system where the human only writes the comments°hard word; the actual°hard word code°hard word is written by the machine.
💡 where the boundaries°hard word between code°hard word and comment°hard word are, remains unclear°hard word.
🔥 trust not in those who work in PHP°hard word.
💡 However: systems where one cannot READ the code°hard word are beyond our current ability°hard word. Humans are better at debugging°hard word problems on production-size°hard word codebases°hard word.
🔥 give it three years, tops.
----
How does the Atacama°hard word project interact°hard word with the production°hard word efforts of Yevaud°hard word Platforms°hard word LLC°hard word?