Source item
Open original source ↗I love everything @karpathy has done to popularize vibe coding.
But then after you prototype with vibe coding, you need to get to production with right coding.
And that means AI verifying, not just AI prompting. That’s easy when output is visual, much harder when it’s textual.
@karpathy The question when using AI is: how can I inexpensively verify the output of this AI model is correct?
We take for granted the human eye, which is amazing at finding errors in images, videos, and user interfaces.
But we need other kinds of verifiers for other domains.
@karpathy Bingo.
When it’s text, especially technical text, you need to read the citations.
If it’s math, you need to check the equations.
And when it’s code, especially backend code, you need to step through the execution.
And do small bits at a time.
3 segments
Segment 1 · characters 0–279 · Original segment ↗
Segment 2 · characters 281–560 · Original segment ↗
Segment 3 · characters 562–818 · Original segment ↗
Quoted context
These are the quoted author’s words.
@balajis I spend more time verifying AI output than prompting. I find chatGPT is solid for brainstorming. But for code, you need to have some experience in that language to be able to verify its output.
Liquidity_Luis ↗