More Than Just a Back Office
For decades, the story of Indian tech, as told in the West, was one of outsourcing. The country’s sprawling tech hubs in Bangalore and Hyderabad were seen as the world’s back office—cost-effective and skilled, but primarily executing on visions created
in Silicon Valley. That narrative is now becoming obsolete. The same forces that built that IT services behemoth—a massive, English-speaking, STEM-educated workforce—are now being redirected toward original creation. The new ambition isn't to service American firms, but to compete with them. This shift is powered by a perfect storm of factors: a colossal domestic market coming online, a world-class digital public infrastructure, and a surge of national confidence. Young engineers who might have once dreamed of a visa to California are now seeing the opportunity to build the next big thing from Mumbai or Delhi.
Meet the New AI 'Unicorns'
This isn't just theory; it’s happening on the ground. Take Sarvam AI. Co-founded by Vivek Raghavan and Pratyush Kumar, two engineers with deep experience in the field, the company raised $41 million in its Series A funding—a record for an Indian AI startup. Their mission isn't just to build another chatbot; it's to create foundational models trained on the vast diversity of Indian languages and cultural contexts. Then there's Krutrim, an AI venture from Bhavish Aggarwal, the founder of Indian ride-hailing giant Ola. Krutrim, which means "artificial" in Sanskrit, recently achieved unicorn status (a valuation over $1 billion) and aims to build the full AI stack, from custom silicon chips to large language models (LLMs) tailored for India. These aren't small-scale application plays; they are audacious, capital-intensive bets on building fundamental AI infrastructure for a nation of 1.4 billion people.
The 'India Stack' Advantage
So, what’s India’s secret sauce? A key ingredient is the “India Stack,” a set of government-backed digital public goods. This includes Aadhaar, a biometric identity system covering nearly the entire population, and the Unified Payments Interface (UPI), which processes billions of digital transactions a month for free. This digital backbone has done two things: it has created an unprecedented amount of structured data about economic and social activity, and it has trained a population to be digitally native. For AI developers, this is gold. Unlike in the West, where data is often siloed in private companies, much of India's foundational data is accessible through open APIs. This allows startups to build and scale solutions for finance, healthcare, and logistics at a speed and cost that would be unimaginable in many other markets. It’s a unique, state-sponsored springboard for innovation.
Challenges on the Horizon
The path forward isn't without obstacles. The most significant challenge is the sheer cost of computing power. Training and running large-scale AI models requires immense GPU capacity, a market dominated by a few American companies like Nvidia. Indian startups are in a global competition for these scarce and expensive resources. Furthermore, while India produces millions of engineers, the number with deep, specialized AI expertise is still limited, leading to a fierce war for talent. Venture capital, while growing, is also still maturing; there’s a need for more domestic investors willing to write large, patient checks for deep-tech research and development, not just quick consumer app wins. Navigating these headwinds will be critical for a homegrown idea to become a global force.
