In a groundbreaking shift in software development, Andrej Karpathy, former head of AI at Tesla and researcher at OpenAI, has coined the term "vibe coding" to describe a new way of coding using large language models (LLMs). This innovative method allows developers to generate code without focusing on the code itself, revolutionizing how startups create technology. The latest batch of Y Combinator, known as W25, exemplifies this trend with a quarter of its startups having 95% of their codebases generated by AI.
Y Combinator managing partner Jared Friedman highlighted this trend, noting the significant role AI plays in the current cohort.
"It’s not like we funded a bunch of non-technical founders. Every one of these people is highly technical, completely capable of building their own products from scratch. A year ago, they would have built their product from scratch — but now 95% of it is built by an AI," said Jared Friedman.
Vibe coding leverages the power of LLMs to automate the coding process, offering a new paradigm for tech startups. Despite its efficiency, this AI-generated code is not without flaws. The first versions of reasoning models reportedly struggle with debugging, making it essential for developers to exercise discernment when using AI-generated code.
Diana Hu emphasized the necessity for developers to possess a discerning eye when engaging in vibe coding.
"You have to have the taste and enough training to know that an LLM is spitting bad stuff or good stuff. In order to do good ‘vibe coding’, you still need to have taste and knowledge to judge good vs bad," she explained.
The shift towards AI-generated code is seen as a lasting change in the tech industry. Garry Tan expressed confidence in the enduring nature of this trend.
"This isn’t a fad. This isn’t going away. This is the dominant way to code. And if you are not doing it, you might just be left behind," stated Garry Tan.