This story is part of CNBC Make It’s The Moment series, where highly successful people reveal the critical moment that changed the trajectory of their lives and careers, discussing what drove them to make the leap into the unknown.
Dhruv Amin and Marcus Lowe left Google and built a startup that was “profitable from Day 1,” operating at a $2.2 million run rate by September 2023, Amin says. They predicted continued success, forecasting monthly revenue to multiply at least five-fold by late-2024.
Then, they shuttered the whole thing to start over from scratch — all due to ChatGPT, says Amin.
Today, Amin and Lowe, both 33, are the co-founders and co-CEOs of Anything, a vibe-coding startup valued at $100 million following an $11 million funding round in September, according to the company. The business was originally known as Create, a marketplace connecting startups to freelance engineers and artificial intelligence-powered coding tools to build websites and apps.
Why blow it up? Amin cites “the ChatGPT moment” in November 2022, when OpenAI debuted the generative AI chatbot. Its leap in capabilities from previous AI models “was a surprise to everybody,” Amin says.
DON’T MISS: The ultimate guide to starting a business—everything you need to know to be your own boss
By spring 2023, Amin and Lowe predicted that AI could eventually approach human developers’ ability to write sophisticated code. If that happened, human engineers might not be needed at all, rendering their business obsolete.
But if they rebuilt their company around AI and the technology didn’t advance rapidly enough, their new model would be a flop. Clients would bolt and the company would run out of money.
After months of anxious deliberation, in October 2023, Amin and Lowe undertook the difficult process of laying off half of their seven-person staff and cutting ties with freelance developers, says Amin: “Within two weeks, we were back to an empty office.”
They first launched an AI-powered tool to build app components like login forms and calendar tools. In April 2025, they released a new product for building entire online businesses, including backend authentication and payments systems, no coding experience necessary — and renamed the company Anything.
“That was actually the moment where it felt like it really took off,” says Amin. Within two weeks of the company’s relaunch announcement, Anything reached a $2 million annualized revenue run rate, he says.
The AI coding industry is “extremely early” in its development, says Amin, as evidenced by overall mixed reviews for vibe-coding services. Still, non-technical users have built functional commercial apps with Anything, including a hair salon owner’s AI stylist mobile app and a dental hygienist’s app for tracking gum health, Amin wrote in a Sept. 29 blog post.
Here, Amin discusses his realization that AI could upend their entire business model, the anxiety of determining whether to start over, and getting the timing right when launching a startup in an emerging industry.
CNBC Make It: You decided to shut down and rebuild in October 2023. How painful of a decision was that, and what was the reaction from your company’s investors?
Amin: We went back to zero, so that was hard. We had to let go of half of the team. We had to shut down the developer marketplace and tell our customers we weren’t going to do this with devs anymore.
At the time, people were still kind of scoffing: “Is this AI thing going to be real? Are you actually going to be able to build real apps [with just AI]?”
We had raised [$3 million] of venture capital. We had promised [investors] that we were going to hit [specific] timelines. Then, we went back to everyone and told them, “Actually, no, we’re going to go back to zero.” Everyone was like, “Are you sure?”
As you and your co-founder tracked AI’s advancements, what were your conversations like?
We talked all the time. We would revisit the position, like, once a month. It took me and Marcus a very, very long time to both come to the conclusion [to start the business over], and it felt like one of those decisions that both of us, as co-founders, needed to agree with.
There’d be times where I was more bullish on the prospects of the [original] marketplace and what we could do with it. There were times where Marcus would be more bullish. And there were times where both of us were like, “This technology is advancing so fast and we’re using it in more and more places, does it just make sense to build a new product?” That was hard.
I think what ended up breaking that logjam for us was: We launched a few prototype builders, just to test and see if there was demand for [code-generation products] where you don’t have to talk to any of us, you just can use it. That gave us some early signal. When we talked to our customers, it felt like the code-gen system had product-market fit.
How confident were you that the underlying generative AI technology would keep advancing exponentially?
Confident, but unconfident on the timescales. That was the scariest part. Through 2024, large language models kind of hit a plateau, to some extent, in capability. It was very unclear through that time how good [that technology] would end up getting. There were definitely [times] where it was like, “Oh, boy. Did we make a mistake in trying to pivot?”
Could we even build something valuable and useful enough that people would want it and pay us for it? We had many opportunities to pivot to a quicker win, which would be: “Let’s make a prototyping tool. Let’s make a design tool, because all [LLMs] can do is components. We’ll sell it to agencies.”
What stopped you?
Marcus and I always had this [mindset] that we’d already gone through one really big pivot trying to align closer to the mission of the company. That pivot would’ve been for naught if we didn’t actually accomplish the mission of the company, which is that people should be able to build real production products, even if they don’t know how to code.
It’s a useful founding story now: This company will stick through it, even if the technology shifts, even if the business model shifts. At the time, it was hard. Now, I can say with some pride: You don’t have to just take my word that we’re serious about what we’re trying to do here. We’ve painted it in blood, sweat and tears.
I expected it would take about a year to get this technology good enough to build real production systems. In reality, it has taken about two and I still think we’re at the start. The sophistication of apps you can build with Anything — what it will look like in a year, or even two years, is just going to be dramatically different.
I hope our hardcore pivots are behind us, at this point. Now, I think we’re just more in the [mindset]: Let’s execute, let’s take advantage of opportunity, and let’s really grow it.
Where would your business be today if you hadn’t shut down the marketplace to focus on AI?
It’s so hard to tell. I wonder if the company would still exist? Would the business we were building then still be attractive to [investors]? I think marketplace models like that, while popular in 2021, have really fallen out of fashion unless you’re more pure AI [focused].
This conversation has been condensed and edited for clarity.
Want to be your own boss? Sign up for CNBC’s new online course, How To Start A Business: For First-Time Founders. Find step-by-step guidance for launching your first business, from testing your idea to growing your revenue.
