In the hypergrowth era of Indian startups, scale was often equated with headcount. But as AI tools mature, the relationship between growth and hiring is loosening.
The employee count is shrinking gradually in the Indian startup ecosystem. The interns were the first to go, but not in a dramatic sweep, or with pink slips or restructuring decks, just quietly replaced by prompts. In the past, if entrepreneur Aditya Goyal wanted to understand how consumers in Jaipur differed from those in Pune, he would dispatch marketing interns into the field. They would fan across the markets, clipboards in hands, asking questions, logging responses into Excel sheets, building dashboards that attempted to map appetite for healthy snacks.
At present, Goyal, the founder of Vishnu Delight, types a prompt to fetch the same data.
Generative Artificial Intelligence is equipped to scan social media behaviour, parsing market chatter and synthesising data points in minutes. “We don’t employ interns for market research anymore,” said Goyal.
“We use the right prompts and decide which city to prioritise, what age group to target, whether it’s office-goers or gym enthusiasts”.
This shift, he estimated, has cut about 50-60% of his data research and marketing costs.
Across India’s direct-to-consumer (D2C) ecosystem- a sector once defined by aggressive hiring and venture-fuelled expansion- founders are rethinking in forming a leaner team. The founders defended, saying: Why hire six when one, assisted by an algorithm, will do?
The tools used in operations are the common ones, mainly OpenAI’s ChatGPT, Google’s Gemini, and Nano Banana. Their impact on India’s startup workforce is only beginning to crystallise.
For Prateek Bhagchandka, founder of food brand MOM Meal of the Moment, the turning point was the photoshoot budget. A high-quality food shoot can cost Rs 1-2 lakh, which is a major outlay for any young brand. Now, many of those images are generated or enhanced through AI tools. “For non-core product visuals, most things can be covered with AI,” he conceded, ” but premium food photography still resists full automation.
Additionally, design briefs that once required coordination between founders, agencies, and freelancers are increasingly replaced by image descriptions fed into generative platforms, said Puru Gupta, founder and operating partner at Proteus Partners. “What once took weeks can now be completed in hours with the help of Nano Banana. The savings are not just financial but structural”.
At fashion brand Libas, AI has quietly reshaped customer communication. Founder Sidhant Keshwani recalled that when WhatsApp messages and push notifications were blasted to entire databases, they were relevant to 20% of recipients only.
However, AI is now capable of segmenting customers by fabric preference, size, price sensitivity, and browsing behaviour with high accuracy. According to Keshwani, the conversion rates have improved, and waste has declined significantly.
On cataloguing, he said, the workforce has reduced dramatically. “Once a 12-14 person operation, it has now shrunk to two people, as the right prompts can help in launching 5,000 SKUs”.
Further, trend forecasting has reduced the waiting game from 4-5 weeks to just 48 hours. “Our in-house AI system predicts, with around 90% accuracy, how many pieces will sell over the three months,” Keshwani added.
He underscored, “If gut is 75%, AI is about 85%…It’s is not perfect, but scans far more than any human can”.
‘ERP cost cutting’
Enterprise resource planning software can cost startups Rs 4-5 lakh annually. For smaller D2C brands, that is a meaningful expense. Goyal sidestepped the subscription model altogether. Using AI layered onto Google Sheets, his team built automated inventory trackers that calculate required quantities and purchasing needs in real time. What once required multiple employees now needs two data-entry and one person overseeing the AI system.
“We are not dependent on human calculations anymore. Everything is system-generated with 90-95% accuracy,” he said.
In supply chains, Bhagchandka’s company is using AI tools to analyse sales data and detect patterns, while upgrading to ERP systems embedded with predictive capabilities. Trend spotting is no longer a quarterly ritual; it is continuous.
‘Fewer hiring’
According to Gupta, the trend is not about mass unemployment but declining hiring intensity.
“Earlier, I might have needed six people. Today, I need one,” he acknowledged that the hiring base has gone down substantially across sectors. In the hypergrowth era of Indian startups, scale was often equated with headcount. But as AI tools mature, the relationship between growth and hiring is loosening.
According to entrepreneurs, routine, data-heavy roles, such as entry-level coding, spreadsheet management, junior content creation, and basic customer service, are most exposed. The grunt work of modern companies is increasingly automated.
Yet founders are careful to frame AI not as workforce replacement but as a filter. Goyal argued that only paperwork-heavy roles are truly vulnerable. Human oversight remained essential to verify outputs, manage systems, and interpret nuance.
Creative work, too, retained a human edge. Bhagchandka admitted that AI cannot yet replicate the artistry of a meticulous shot food photograph, while Keshwani said that innovation still demands taste and instinct. But the bar has shifted.
If fewer people are being hired, those who are must bring something different to the table. Bhagchandka said that the ability to craft effective prompts and work fluidly with AI tools is fast becoming a core competency. “Willingness to learn AI and implement it is critical,” he added.
