People with no knowledge of coding or computer languages can use vibe coding using plain English command with AI-assisted software development.
AI is reshaping the software development lifecycle (SDLC), with machines now writing thousands of lines of code once handled by engineers.
In response, information technology services companies are increasingly seeking talent from non-technology backgrounds to tap into their domain expertise.
While these hires are not trained engineers and cannot write code themselves, companies are unconcerned because code-generation (often referred to as code-gen) startups can autonomously handle coding tasks.
At Cognizant, nearly 30 per cent of code is now machine-generated, up from 20 per cent earlier this year, said Ravi Kumar, its CEO.
"We will continue to need deep programmers who build platforms and write AI algorithms. But as SDLC evolves, we will also hire people with other domain expertise and train them for the future.
"We could be hiring chartered accountants and MBAs in finance to agentify the finance and accounting functions of a company," Kumar said during an earnings press meet last month.
As functions such as finance, accounting, legal, and supply chain management become automated -- or gain AI agents to assist employees -- IT companies are likely to recruit people with deep industry knowledge who can provide recommendations, not just execute tasks.
Executives like Kumar are confident thanks to a concept called 'vibe coding', coined by Andrej Karpathy, cofounder of OpenAI and former head of AI at Tesla.
It allows people with no coding knowledge to write code using plain English commands.
Startups such as US-based Cursor and Windsurf -- acquired by Google for $2.4 billion -- have become investor darlings in the emerging code-gen sector.
Google CEO Sundar Pichai said in April that 'well over 30 per cent' of the company's code is now AI-generated.
Amazon CEO Andy Jassy noted in a LinkedIn post last year that the company's generative AI assistant for software development, Amazon Q, helped save the 'equivalent of 4,500 developer-years' when upgrading applications to Java.
"You have to start thinking of disciplines differently -- a broader spectrum of skills and capabilities rather than just domain STEM (acronym for science, technology, engineering, and mathematics) skills," Kumar added.
"There are also areas, such as legal, where classical software has barely touched. There is very little software in legal, but it's the right candidate for agentification," Kumar added.
Experts say companies will hire based on three key parameters going forward: Tech expertise (knowledge of AI skills), domain expertise (banking, retail, life sciences), and client-specific expertise, which involves understanding the customer better through agentic AI.
This approach requires deeper contextual knowledge, or 'contextual engineering', where engineers must understand not just what they are building with, but also what they are building for.
The more contextual awareness an engineer or manager has, the better they can monitor code and agents, preventing errors before they occur.
Anand Birje, CEO of engineering services company Encora, predicts that vibe coding tools will create a new generation of coders and developers over the next three to four years.
"We actually feel we're going to hire more from both engineering schools and management, commerce, and humanities backgrounds. With contextual engineering, we will need all three disciplines and less of pure development," Birje said.
Sangeeta Gupta, senior vice-president and chief strategy officer at Nasscom, told Business Standard that the IT industry will increasingly need individual contributors.
"It's more about who has expertise in a particular theme, technology, and domain. These individual contributors will build the right partnerships, provide recommendations, and not manage people."
What is vibe coding?
People with no knowledge of coding or computer languages can use vibe coding using plain English command with AI-assisted software development.
It was a term coined by Andrej Karpathy, co-founder of OpenAI and former head of AI at Tesla.
Firms such as Cursor and Windsurf are some examples of firms using vibe coding.
Feature Presentation: Ashish Narsale/Rediff