Post

I have different opinions. I review LLM-generated code far more often than I like. It's the worst part of my job, and probably my career. I really want to see a John Henry-style humans vs AI code-off. With a 6-month follow-up to see how that genAI code is doing.