‘At what cost?’ Sridhar Vembu flags AI’s double-edged impact on coding jobs & engineers

Zoho co-founder and Chief Scientist Sridhar Vembu has ignited a fresh debate on the future of software engineering, warning that the rapid adoption of artificial intelligence could upend the traditional talent pipeline even as it delivers dramatic productivity gains. 

In a recent post on X (formally twitter), Vembu described AI as a “double-edged sword” for the technology workforce. While advanced AI tools are making senior engineers and architects far more productive, he cautioned that the same tools could sharply reduce the need for junior engineers — potentially breaking the long-established pathway through which future architects are trained. 

“AI makes senior architects more productive and reduces the need for junior engineers,” Vembu wrote. He noted that effective use of AI requires deep understanding of both business requirements and the technology stack, skills typically acquired through years of hands-on experience. “But if we don’t have junior engineers, we don’t get to train the next generation of architects,” he added, posing a fundamental question about how expertise will be cultivated in an AI-driven industry. 

The comments come as Zoho continues to experiment with AI coding assistants, including Anthropic’s Claude Opus 4.5. Vembu has long spoken about AI’s transformative potential in software development, but his latest remarks underscore a growing concern across the industry: whether efficiency gains could come at the cost of long-term workforce sustainability. 

Vembu’s post triggered wide-ranging responses online, with technologists and engineers debating how job roles might evolve. Some suggested redefining junior roles to focus on AI supervision, prompt engineering, and system-level thinking, while others warned that sidelining early-career engineers could lead to acute talent shortages in the years ahead. 

Engaging with users in the thread, Vembu offered advice to those entering the field. He argued that aspiring engineers should aim to become “strong domain experts in some specialised domains,” calling it the safest path in an era where routine coding tasks are increasingly automated. Core technical roles involving algorithms, protocols, and foundational systems will continue to exist, he said, but in smaller numbers. 

Vembu also took a broader, more philosophical view of AI’s societal impact. Responding to a comment that framed AI as a way to escape the pressures of urban tech life, he predicted a shift in what society deems economically valuable. Activities such as caring for the environment, tending to children and the elderly, and offering spiritual guidance could, he suggested, become more valuable than writing code as AI reshapes productivity and labour markets. 

His reflections were reinforced by a striking example from within Zoho itself. A day earlier, Vembu shared how a senior engineer in the company’s R&D team had independently built an advanced assembly and machine-code security tool in just one month — a task that would traditionally require a team of three to four engineers working for at least a year. 

According to Vembu, he was unaware the engineer was even developing the tool until it was demonstrated. “I was blown away by the depth and breadth of the tool,” he wrote, describing the project as a powerful illustration of AI-enabled productivity. Notably, the engineer had previously been sceptical of AI-generated code, underscoring how hands-on experimentation can rapidly change perceptions. 

“This is how we learn in Zoho — we let smart people experiment and find new pathways,” Vembu said.