Why India's Chip Founders Didn't Exist Before 2022, and What Actually Changed
One founder's quote reveals what funding data can't: an entire category of entrepreneur was created by two policy decisions, not market demand.

Ask a semiconductor founder when Indian chip-design entrepreneurship actually started, and the honest answer isn't a technology milestone. It's a policy date. Hareesh Chandrasekar, co-founder and CEO of Agnit Semiconductors, put it in six blunt words: before 2021-2022, nobody was even interested in semiconductors as a startup category at all. Not "interested but underfunded." Not "interested but cautious." Genuinely absent as a founder path, the way nobody starts a company today betting on fax machines.
Pre-2021-2022, nobody was even interested in semiconductors.
That's a striking thing for a founder to say about his own category, and it's worth taking at face value rather than treating it as an exaggeration for effect. Semiconductor design requires years of specialized technical training, deep capital, and a tolerance for multi-year development cycles before any product ships, exactly the profile of founder that venture-backed India spent the previous decade actively selecting against. The dominant founder archetype through the 2010s and into the early 2020s built software, fast, cheap, and quick to iterate. A chip-design startup is the structural opposite of that on every axis: slow, capital-intensive, unforgiving of iteration mistakes. There was no market failure here. There was simply no founder pool willing to make that trade, because nothing was rewarding them for trying.
What actually changed the incentive
Chandrasekar is specific about the mechanism, and it isn't vague policy atmosphere. Two concrete government programs, the India Semiconductor Mission and the Design Linked Incentive scheme, were what he credits with making chip-design startups a viable path to founders who own their own IP-designed products, rather than remaining engineers inside someone else's multinational chip-design center. That distinction matters more than it sounds. India has had strong chip-design talent for years, roughly 20% of the world's chip-design workforce sits inside the country. What it never had was a reason for that talent to leave a stable multinational job and start something of its own.
The DLI scheme's actual mechanism is worth spelling out, because it's more specific than "government support for startups" implies. It directly subsidizes the cost of chip design and IP development for companies choosing to own their own semiconductor intellectual property rather than build it as work-for-hire inside a larger company's design center. That's a precise financial lever aimed at exactly one decision: an engineer's choice between staying employed at a stable multinational salary or taking on the multi-year, no-revenue-until-shipping risk of owning IP themselves. Before the scheme existed, that math simply didn't favor leaving. The subsidy didn't make chip design easier or the technology less risky. It made the specific financial gap between "employed engineer" and "IP-owning founder" survivable long enough for someone like Chandrasekar to actually cross it.
The results are showing up in real numbers, even if the category is still small by absolute standards. Indian deep-tech companies, semiconductors among them, raised roughly $1.06 billion across 137 equity rounds in the twelve months to July 2025, more than double the same period a year earlier, according to Tracxn data. The full 2024 calendar year saw deep-tech funding of $1.6 billion, a 78% year-on-year increase. By August 2026, Tracxn's broader deep-tech count had grown to 8,842 startups total, with 448 of them founded in 2026 alone, though only seven have reached unicorn status and just 373 have secured Series A or later, a reminder that "founded" and "funded past the earliest stage" remain two very different bars in this category.
The gap between approval and production
None of this means the policy machinery is running cleanly. As of April 2026, the government had approved 12 semiconductor manufacturing projects under the India Semiconductor Mission, backed by a planned outlay exceeding ₹1.64 trillion, and only two of the first ten approved projects had actually begun commercial production. That's the honest state of things: approvals move fast, ribbon-cuttings happen on schedule, and actual manufacturing output lags well behind both. Chandrasekar's own founder cohort is proof the incentive structure successfully created new companies. Whether that same structure gets those companies from "founded with IP ambitions" to "shipping chips at commercial volume" is a different, harder question, and the two-out-of-ten production rate is the number that should temper any premature victory lap.
What this means beyond one founder's quote
The real significance of Chandrasekar's line isn't the nostalgia value of remembering a category that didn't used to exist. It's that it demonstrates something India's policymakers evidently already understood and bet on correctly: founder behavior in a capital-intensive, high-risk category responds directly and quickly to a specific subsidy mechanism, not to general enthusiasm, summit speeches, or funding-total headlines. The DLI scheme didn't make semiconductors interesting. It made the specific financial math of starting a chip company survivable for a founder who previously had no reason to try. That's a narrower, more useful lesson than "India is having a deep-tech moment," and it's the one worth remembering the next time a policymaker wants to know whether a subsidy actually changed founder behavior or just funded a press conference: ask the founders who showed up only after the incentive did, and take their answer as literally as Chandrasekar's.



