Can Vibe Coding Mark The End Of SaaS Dominance?

Advertisement
Read Time: 4 mins
India is home to more than 1000+ SaaS companies. Representational (Image By Freepik)

It is around 3PM on a work day and a developer is sitting at their work/gaming desk, thinking of creating a new app. The developer opens the AI app of their choice, puts in the prompt and the machine generates the code needed for the project. Gone are the days when this developer sat and coded for days on end, before testing the app. Now, they can skip to the good part.

What we described above is referred to as "vibe coding", a term coined by computer scientist Andrej Karpathy, and it is spreading fast. But does the prevalence of vibe coding mean that Software as a Service (SaaS) companies have seen the end of their dominance? The market is divided.

Advertisement

Vibe-coding platforms have strengthened engineering infrastructure, says Khadim Batti, co-founder and CEO, Whatfix. Such platforms handle large portion of boilerplate, repetitive code, and even initial drafts of testing frameworks, which previously consumed a lot of time, he said.

“By offloading this work to AI, our teams can now focus on higher-value activities such as architecture, design decisions, optimisation, and solving complex business problems," Batti said. But these platforms are not affecting overall business yet.

Advertisement

At Zoho, business seems to be booming, despite vibe coding. Praval Singh, vice president for marketing and customer experience said the company has reported a customer growth of 50% this year. Zoho has been working on its AI stack for close to a decade, where the past two years have seen AI capabilities being passed across the product lineup.

The company has also unveiled Large Language Model (LLM) to help businesses in easing their workflows. The SaaS firm also offers Zoho Creator, a low-code app development platform, featuring many AI capabilities.

Advertisement

AI and vibe coding have a long way to go and are still in its nascent stages. “Much of what we see today is promising, but still experimental. AI generated code still needs to be checked by a professional coder to catch hallucination, similarly, vibe coding is riddled with backend opacity which limits full control,” says Singh.

Though, last 12-18 months saw enterprises making significant investments in AI copilots and AI agents, but it's the ROI that is still taking a hit, said Batti.

Moreover, tasks associated with financial management revolves careful handling of edge cases and data integrity. This is something which AI finds tough to handle, Singh said.

"Some of the leading platforms today have been built over years of R&D to tackle the needs of the business and offer a solution which isn't limited or has a complex learning curve," he said. "...but one that can be tailored to their unique uses and understands business complexity.”

Vibe coding cannot result in solutions like email or spreadsheet, requiring years of R&D, according to Singh. Also, when compliance requires regulatory oversight, it cannot be substituted with AI.

Advertisement

According to Sankar Lagudu, co-founder and COO at Responsive.io, human expertise and oversight are still needed in critical sectors like financial services, healthcare, and government.

Loading...