A screenshot of the website for Chloe Samaha’s startup BOND, an earlier version of which was almost entirely vibe coded.
bondapp.io/Screenshot by NPR
hide caption
toggle caption
bondapp.io/Screenshot by NPR
Chloe Samaha wasn’t trained to write software. But she and her partner at their San Francisco-based startup BOND got a working version of a new online productivity manager and website up and running in less than a day.
“He was on his way back from a ski trip and built the entire back end … in six hours. And I built the front end in, like, an hour-and-a-half, and we just had a functional product,” Samaha said.
They did it mostly by “vibe coding” — using fast-evolving artificial intelligence chatbots, as well as other new AI tools, to write the software for them.
It helped Samaha go from a concept for what she calls “an AI chief of staff for CEOs and busy execs” to a prototype and then a product on the market with lightning speed.
It also highlights advances in AI that are opening up possibilities for creators and shaking up the world of software engineering.
Samaha, 21, said BOND’s product, called “Donna” (named after a character on the TV series Suits), taps into users’ data from various platforms, like email, calendars or Slack, and uses AI to give instant answers to questions about things like progress on projects and team performance.
The firm recently got a $500,000 investment from the venture capital firm and tech incubator Y Combinator.
Tom Blomfield is a group partner there. He said the term “vibe coding” was coined by OpenAI co-founder Andrej Karpathy earlier this year in a tweet.
“It really caught on, this idea that people are no longer checking line by line the code that AI is producing, but just kind of telling it what to do and accepting the responses in a very trusting way,” Blomfield said.
And so Blomfield, who knows how to code, also tried his hand at vibe coding — both to rejig his blog and to create from scratch a website called Recipe Ninja. It has a library of recipes, and cooks can talk to it, asking the AI-driven site to concoct new recipes for them.
“It’s probably like 30,000 lines of code. That would have taken me, I don’t know, maybe a year to build,” he said. “It wasn’t overnight, but I probably spent 100 hours on that.”
Blomfield said he expects AI coding to radically change the software industry. “Instead of having coding assistance, we’re going to have actual AI coders and then an AI project manager, an AI designer and, over time, an AI manager of all of this. And we’re going to have swarms of these things,” he said.
Where people fit into this, he said, “is the question we’re all grappling with.”
In 2021, Blomfield said in a podcast that would-be start-up founders should, first and foremost, learn to code. Today, he’s not sure he’d give that advice because he thinks coders and software engineers could eventually be out of a job.
“Coders feel like they are tending, kind of, organic gardens by hand,” he said. “But we are producing these superhuman agents that are going to be as good as the best coders in the world, like very, very soon.”
Others are less sure that the future for coders is so grim.
Tools that are now available for vibe coding are lowering the barrier to entry for software development. For many designers, it’s a boon. They can now try lots of things quickly, rather than having to hand their ideas off to software engineers to create prototypes, one by one.
“For every idea, there are 10 ideas that are not explored today because there simply isn’t the time,” said Yuhki Yamashita, chief product officer at Figma, a leading design software firm.
Earlier this month, Figma rolled out an AI-enabled product that effectively lets designers vibe code. “So we think that more ideas will be explored and more ideas will be validated much, much faster,” Yamashita said.
Thomas Dohmke, CEO of the software development platform GitHub, which is owned by Microsoft, said AI is a “huge opportunity for software developers.” He sees a future where software developers may be like conductors presiding over orchestras of AI coding agents.
“The question is, from my perspective, not so much: Is the job going to go away? But: How is the job going to evolve?” he said.
“There is going to be a commoditization of simple tasks that AI can do, and there always has been. If you can automate it, developers will do that because they have so much work to do,” he continued.
Adam Resnick, a research manager with the tech consultancy IDC, said software developers who might be worried about their jobs still have some runway ahead of them — even if things are progressing quickly.
“The vast majority of developers are using AI tools in some way,” he said. “And what we also see is that a reasonably high percentage of the code output from those tools needs further curation by people, by experienced people.”
And that’s a job that AI can’t do, he said. At least not yet.
