A conversational take on the importance of human taste and judgment when using AI-generated code. The core argument is that AI models are trained on the entire corpus of human-written code — good and bad — and developers must act as 'tastemakers' to constrain AI output to code they'd actually want to maintain. The analogy drawn is to Ruby on Rails, which succeeded not just because of the language but because of strong opinionated design choices.
•1m watch time
50 Impressions