We’re living in a time when individuals with no formal engineering background are creating and selling apps using platforms like Lovable, Replit, and Windsurf.
Some people recoil at what’s now being known as ‘vibe coding’, while others can’t stop talking about it.
However, the truth is, it works—it is entirely possible to build something independently and earn profits from it without being a fully fledged developer.
That said, not everyone who uses these tools has entrepreneurial endeavours. Many individuals simply want to learn the latest tools in exchange for a stable job at an established company.
So the question is: does vibe coding help pave the way in any meaningful way?
‘Vibe Coding Experience is Non-Negotiable’
In recent weeks, several startups have hired developers specifically for their vibe coding skills.
For instance, DV8 Infosystems, a US-based IT services company, was seeking an ‘AI-powered developer’ on LinkedIn, specifically stating, “No traditional computer science degree required—just the right mindset.”
Then there’s Domu Technology, a Y Combinator-backed San Francisco-based startup that automates debt collection calls for banks. The company is offering up to $150,000 annually plus equity for a role titled ‘vibe coder-AI engineer’. The job posting makes it clear: “At least 50% of the code you write right now should be done by AI; vibe coding experience is non-negotiable.”
Moreover, several other startups are seeking experienced engineers skilled in AI-enabled coding tools, offering salaries of up to $200,000 per year. Notably, AIM came across an individual seeking a vibe coding coach on Upwork, offering up to $35 per hour.
Founders are increasingly recognising the advantage of hiring developers who are fluent in tools like Cursor and Windsurf.
A startup founder, on condition of anonymity, told AIM that they are looking to hire a vibe coder with at least a year of experience in programming, solely because these tools are especially valuable when working across multiple frameworks and programming languages with limited resources.
They further stated that developers with basic proficiency in just one language, paired with these tools, are now able to contribute code in languages they haven’t formally been trained in, making them far more versatile and productive in lean teams.
John Leimgruber, a developer with over a decade of experience in building software, stressed the necessity of proven development experience. He told AIM that while he resonates with the advantages that coding brings, concerns regarding security, front-end or back-end API integration, and the database layer remain.
Leimgruber also pointed out that developers still need a solid foundational understanding to know which libraries to use, how to handle concepts like async and concurrency, and how to identify bottlenecks across CPU, RAM, input/output, network, and disk.
“A joke I heard recently is, ‘vibe coding allows a single developer to generate the technical debt of 50 developers’,” Leimgruber said.
In conclusion, while many startups are leaning into vibe coding, the general consensus is that solid experience with handling code, understanding how to debug errors, and a background in working with production-level code still matter.
Leimgruber also noted that real-world development environments involve far more nuance and context than AI tools can currently handle.
Most companies rely on legacy databases and systems, which introduce a layer of complexity that goes beyond what AI interfaces are built to manage. He also stated that AI might be useful for building greenfield applications, but it falls short when dealing with tasks like migrating existing systems. “AI is probably not so good to replace the operations team for a legacy product.”
Even in the case of Domu Technology’s ‘vibe coder – AI engineer’ role, the job posting still called for over three years of experience and a proven track record of delivering scalable products to end users.
It turns out, vibe coding can get you there.
No Better Way to Learn Code
In an interaction with AIM, Karan MV, the director of international developer relations at GitHub, shared his perspective on the demand for AI-enabled coding tools.
Karan believes that with the advent of AI, the demand for software developers is increasing. He attributed the trend to businesses having a “lot of backlogs” and wanting to accelerate the development process.
“What we envision is for people to start learning coding with the help of AI,” he said, indicating that these AI tools can assist individuals in upskilling at an unprecedented level and, subsequently, empower them with the knowledge to address real problems within larger organisations.
Besides, the current advancements in the AI app ecosystem suggest that it is a great time to hop on to these tools. For example, Google’s Gemini 2.5 Pro Experimental model is being widely regarded as the best model for coding. Turns out, it is available for free users on the Gemini app.
A few months ago, GitHub also announced that GitHub Copilot is available for free and integrates vibe coding abilities into the platform.
