{"id":927837,"date":"2026-07-11T12:53:46","date_gmt":"2026-07-11T12:53:46","guid":{"rendered":"https:\/\/www.europesays.com\/us\/927837\/"},"modified":"2026-07-11T12:53:46","modified_gmt":"2026-07-11T12:53:46","slug":"the-cliff-that-sends-indias-founders-to-san-francisco","status":"publish","type":"post","link":"https:\/\/www.europesays.com\/us\/927837\/","title":{"rendered":"The Cliff That Sends India’s Founders To San Francisco"},"content":{"rendered":"
  • India has learned to fund startups into existence. But between the first cheque and a future listing lies a chasm that deep-tech founders cannot cross at home, so the hardest of them cross an ocean instead.<\/li>\n

    Sometime in the early months of 2025, in a modest Bengaluru laboratory, a young engineer named Anagha Rajesh managed something that until recently had lived only inside the research wings of Microsoft and a handful of American universities. She wrote digital data into the DNA of living bacteria, then read it back out again, intact.<\/p>\n

    Her company, BioCompute, had been running for barely a year, assembled from a graduate-school frustration with cloud-storage bills, a grant, an incubator bench, and a non-dilutive cheque from Nikhil Kamath’s WTFund, for which she had beaten all but eight of more than 1,500 applicants. Measured against the global field her progress had been almost implausibly cheap; measured against the appetites of Indian venture capital, close to invisible.<\/p>\n

    By the middle of 2026 she had decided to leave, taking her company and her seed round to San Francisco.<\/p>\n

    She is not, on the evidence, fleeing indifference. She is walking off the edge of a cliff that India’s own ecosystem has learned to name.<\/p>\n

    The country has become genuinely good at bringing companies into being, writing the grants, the accelerator places, the seed cheques that turn a graduate-school idea into a going concern. What it has not built is the far side of the climb: the second, third and fourth rounds, growing steadily larger, that carry a company across a decade towards a public listing.<\/p>\n

    Something like 85 per cent of India’s seed-funded ventures never reach even a Series A, a drop-off steep enough that founders call it the graduation cliff. For most startups it is survivable, because a consumer app can reach revenue in a year or two and haul itself up the other side.<\/p>\n

    For a hard-science company it is very nearly fatal, and the story of why Anagha Rajesh is on a flight to California is, at bottom, the story of a cliff that is far steeper for some founders than for others.<\/p>\n

    \"Around<\/p>\n

    Around 85 per cent of India’s seed-funded startups never reach a Series A \u2014 survivable for a consumer app that reaches revenue in a year, near-fatal for a decade-long science bet.<\/p>\n

    Why her climb is so long<\/strong><\/p>\n

    To see why deep tech falls so hard, begin with what she is actually building, because the technology carries the argument rather than merely setting the scene for it.<\/p>\n

    Every file is, at bottom, a long string of ones and zeros, and DNA stores information too, in a four-letter chemical alphabet; the idea is to translate the binary into sequences of those bases so that a file becomes, quite literally, biology.<\/p>\n

    DNA is almost unimaginably dense (one estimate cited by BioCompute holds that a drive the size of a teaspoon could store all the world’s data), durable enough to be read after thousands of years, and at rest it draws no power at all.<\/p>\n

    The obstacle is cost. Reading DNA has grown cheap with remarkable speed, sequencing costs having fallen faster than Moore’s law over two decades, but writing it has not; building DNA base by base through chemical synthesis remains slow and expensive, which is why every serious effort treats the technology as cold archival storage, written once and read almost never.<\/p>\n

    The first commercial DNA data centre is still reckoned to be five to ten years away.<\/p>\n

    BioCompute’s wager is to attack that cost wall from an unusual angle, storing data inside the genomes of living bacteria, edited in place rather than synthesised from scratch. It is inventive and unproven, the honest description of most things at this frontier.<\/p>\n

    That decade-long timeline is what makes her cliff so nearly fatal, and why everything now turns on whether someone at home will fund the long climb up the other side, and if not, why not.<\/p>\n

    Why the money won’t follow her over<\/strong><\/p>\n

    The tempting explanation, now most loudly attached to her departure, is that India simply lacks the money.<\/p>\n

    Vyom Bhatia, an investor who has campaigned in public to keep BioCompute in the country and to raise its seed round himself, argues that the country is not poor at all: depending on the list, it counts more than 250 dollar billionaires. What it lacks, in his telling, is not capital but conviction, not money but awareness.<\/p>\n

    He reaches, as such arguments reflexively do, for the founding of OpenAI: in 2015 a few wealthy men pledged a billion dollars to a lab meant to serve humanity rather than profit, and look what grew from it. India has the billionaires; it needs only a handful to make the same bet. Bring our people home, he writes. Any takers?<\/p>\n

    The instinct is generous, but it locates the whole problem in other people’s inattention and none of it in the structure of the thing itself.<\/p>\n

    The money is genuinely there (Forbes counted 284 Indian billionaires in 2025, third in the world behind the United States and China), so the question is never whether the rupees exist but where they consent to travel, and here Indian venture capital has a pronounced, well-documented shape.<\/p>\n

    The annual surveys by Bain and by the Indian Venture and Alternate Capital Association show, year upon year, that consumer technology, software-as-a-service and fintech together absorb more than 60 per cent of all venture funding, with consumer tech alone drawing roughly $5.4 billion in 2024 out of about $13.7 billion. This is rational: those businesses enjoy huge domestic markets, quick revenue and, increasingly, real exits, exactly the companies for which the cliff is shallow.<\/p>\n

    \"Consumer<\/p>\n

    Consumer tech, SaaS and fintech together absorb more than 60 per cent of Indian venture funding; deep tech drew about $1.6 billion in 2024, a sliver of the roughly $13.7 billion total.<\/p>\n

    Deep technology, by contrast, remains a small if swiftly swelling sliver, with Tracxn putting Indian deep-tech funding at about $1.6 billion in both 2024 and 2025 and cumulative funding since 2016 approaching $28 billion.<\/p>\n

    Ask the people who actually deploy that capital what stops them, and hardly any cite ignorance; they describe a structural hole in the financing chain. The binding constraint, as Vishesh Rajaram of the deep-tech fund Speciale Invest frames it, has historically been the depth of funding at Series A and beyond.<\/p>\n

    India has learned to provide early grants and seed cheques, several of which Ms Rajesh collected, but not the large domestic funds able to carry a science company across a decade and several rounds towards a distant listing. The IVCA’s own president concedes as much: founders in this space long looked to Silicon Valley, because the funds able to understand a seven-year science bet scarcely existed.<\/p>\n

    A founder who reaches the cliff-edge and still needs hundreds of crore for the climb finds few hands extended at home, and turns abroad out of arithmetic, not any deficit of patriotism.<\/p>\n

    The OpenAI analogy, examined rather than invoked, only sharpens the point. The billion dollars of 2015 was overwhelmingly a pledge rather than a transfer: filings show only around $133 million had been received by 2021, with Elon Musk’s own contribution under $45 million.<\/p>\n

    Nor did the noble structure survive contact with reality; by 2019 OpenAI had grafted on a for-profit arm because its costs would dwarf anything philanthropy could supply, and it is today a public-benefit corporation that has filed to go public at a valuation in the trillions.<\/p>\n

    The favourite example of patient, humanity-serving capital thus turns out to be the clearest demonstration of how fast that posture buckles at the frontier. A billionaire’s act of conscience can fund the first step off the cliff; it cannot fund the years of descent. Philanthropy can light the fuse; it rarely sustains the burn.<\/p>\n

    Why so few even reach the edge<\/strong><\/p>\n

    If the wrong shape of capital makes the cliff impossible to cross, a second force ensures that few Indian deep-tech firms ever reach its edge at all, and it is the one no individual’s awareness can touch.<\/p>\n

    India does not, in any meaningful sense, fund research. Its gross spending on research and development hovers around 0.65 per cent of GDP, against roughly 2.4 per cent for China and 3.5 per cent for the United States, and the proportions are inverted, with the state supplying close to two-thirds and private industry barely a third, where in the advanced economies industry carries the bulk.<\/p>\n

    A nation spending a fifth of its rival’s share on the very science from which deep-tech companies are grown does not have an awareness problem among its billionaires; it has a foundation problem in its laboratories, universities and corporate balance sheets. The graduation cliff is what founders see; the shallow science base is why the population at its edge is so thin to begin with.<\/p>\n

    \"India<\/p>\n

    India spends about 0.65 per cent of GDP on research and development, against 2.4 per cent for China and 3.5 per cent for the United States \u2014 and unlike them, the state, not industry, carries most of it.<\/p>\n

    There is a further inconvenience for the brain-drain framing: the dominant capital story of the past two years has run the opposite way. Indian startups have not been fleeing but coming home, at considerable expense, in a movement the ecosystem calls the reverse flip.<\/p>\n

    PhonePe set the precedent in 2023, paying close to a billion dollars to shift its domicile from Singapore back to India; since then Meesho has paid roughly $300 million in tax to repatriate, Razorpay around $150 million and Groww about $160 million, with Zepto, Dream11 and Flipkart in the same procession. The magnet is the maturing of India’s own capital markets and the abolition of the angel tax.<\/p>\n

    But note what kind of companies these are: every one is a consumer or fintech business whose customers are Indian and whose exit is an Indian IPO, returning home precisely because home is now the best place to sell to Indians and list for Indians.<\/p>\n

    That offers almost nothing to a pre-revenue, globally addressed, decade-from-any-IPO deep-tech venture whose buyers are the world’s hyperscale data centres. The reverse flip is real and heartening, and it sails clean past Anagha Rajesh.<\/p>\n

    \"The<\/p>\n

    The tax bills consumer and fintech firms have paid to redomicile to India, from PhonePe’s roughly $1 billion down \u2014 a homecoming driven by Indian customers and Indian IPOs that offers little to pre-revenue deep tech.<\/p>\n

    The bet India has placed<\/strong><\/p>\n

    The most striking feature of the appeal to rouse the country’s billionaires is that India has, in effect, already begun to build the far side of the cliff, and from an unexpected quarter.<\/p>\n

    On 3 November 2025 the government launched a Research, Development and Innovation Fund with a corpus of Rs 1 lakh crore, roughly $12 billion, and its architecture is patient capital by design: the money reaches the Anusandhan National Research Foundation as a fifty-year, zero-interest loan, to be deployed as long-tenor, low-cost finance for projects at the scaling stage, with a dedicated Deep-Tech Fund of Funds built in and Rs 20,000 crore earmarked for the first year alone.<\/p>\n

    This is, whatever its flaws, the state placing exactly the long, risk-tolerant bet the parable demands, at a scale no individual billionaire would contemplate; Kris Gopalakrishnan, the Infosys co-founder, has called it a potential energiser for deep-tech capital.<\/p>\n

    Nor does it stand alone. A private coalition, the India Deep Tech Alliance, has gathered more than a billion dollars from the likes of Accel, Blume Ventures and Qualcomm Ventures, with Nvidia advising, aimed squarely at the Series A-and-beyond gap founders such as Ms Rajesh fall into.<\/p>\n

    Startup recognition for deep-tech firms has been extended to twenty years to match their gestation, and the Technology Development Board and the biotech agency BIRAC each have Rs 2,000 crore to deploy. As Siddarth Pai of 3one4 Capital observes, the patience that was for so long missing has begun to arrive, with investors increasingly untroubled by waiting ten or fifteen years.<\/p>\n

    The caveat is that announcing capital and deploying it well are different achievements, and India’s record at the second is, plainly, mixed.<\/p>\n

    Critics note that the foundation’s board is heavy with bureaucrats and academics and light on industry and global expertise, and it will be judged not by its press releases but by whether the money reaches, at arm’s length from the state, the founders who would otherwise leave.<\/p>\n

    Whether the bet pays out is a question of execution, exactly what India has most reliably fumbled. But the claim that no one had built any bridge across the cliff is no longer true, if it ever was.<\/p>\n

    The far side is in San Francisco<\/strong><\/p>\n

    Which leaves the hardest part of the problem, the part money alone does not solve: even a fully funded crossing does not explain why the far side of the cliff, for a company like BioCompute, currently sits in California.<\/p>\n

    Capital is only one of the things a ten-year science bet needs. The others are a dense cluster of synthetic-biology and microfluidics talent, lead customers willing to pilot an unproven storage medium, and a credible exit at the end of the decade, and that cluster still sits most thickly in San Francisco, which India’s improving but still-shallow deep-tech stack cannot yet match in one place.<\/p>\n

    She is not leaving because no Indian billionaire has heard of her; she is leaving because everything a founder needs on the far side of the cliff is, for now, on the other side of an ocean.<\/p>\n

    There is a line, relayed by Mr Bhatia, that Ms Rajesh is said to have told him about the Indians already working in America whom she means to hire: that they do not want to raise their children there, and want, in the end, to come home.<\/p>\n

    It is almost certainly true, and perhaps the most important sentence in the story, because the talent is not lost to sentiment or disloyalty but held abroad by infrastructure. Those people will come home to a working laboratory, a market that will buy from them, and an exit that rewards a long bet, not to a tweet, however heartfelt.<\/p>\n

    That leaves the work in the unglamorous register no farewell video can dramatise: raising the share of national output India spends on research and pressing private industry to shoulder more of it; funding the growth-stage vehicles that write the second, third and fourth cheque, not merely the first; making the government a real customer through procurement; and engineering exits that repay a decade of patience.<\/p>\n

    The state has, belatedly and imperfectly, begun several of them at once, which is more than its critics expected and less than the moment requires.<\/p>\n

    The teaspoon of DNA that could one day hold all the world’s data will be built somewhere in the coming decade, by someone.<\/p>\n

    Whether that somewhere is Bengaluru or Berkeley, and whether the someone is an Indian who stayed or an Indian who left and later sold her company back to the country at a markup, will not be settled by how many billionaires were made aware of her name.<\/p>\n

    It will be settled by whether India proves willing to do the slow, expensive and profoundly unspectacular work of building the far side of the cliff, quickly enough to keep the next founder from stepping off it. That is a far harder ask than awareness. It is also the only one that has ever worked.<\/p>\n","protected":false},"excerpt":{"rendered":"India has learned to fund startups into existence. But between the first cheque and a future listing lies…\n","protected":false},"author":3,"featured_media":927838,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":"","_share_on_mastodon":"0"},"categories":[16],"tags":[362562,64,607,374270,67,132,68],"class_list":["post-927837","post","type-post","status-publish","format-standard","has-post-thumbnail","category-entrepreneurship","tag-biocompute","tag-business","tag-entrepreneurship","tag-indian-founders","tag-united-states","tag-unitedstates","tag-us"],"share_on_mastodon":{"url":"https:\/\/pubeurope.com\/@us\/116901446452893618","error":""},"_links":{"self":[{"href":"https:\/\/www.europesays.com\/us\/wp-json\/wp\/v2\/posts\/927837","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.europesays.com\/us\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.europesays.com\/us\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/us\/wp-json\/wp\/v2\/users\/3"}],"replies":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/us\/wp-json\/wp\/v2\/comments?post=927837"}],"version-history":[{"count":0,"href":"https:\/\/www.europesays.com\/us\/wp-json\/wp\/v2\/posts\/927837\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/us\/wp-json\/wp\/v2\/media\/927838"}],"wp:attachment":[{"href":"https:\/\/www.europesays.com\/us\/wp-json\/wp\/v2\/media?parent=927837"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.europesays.com\/us\/wp-json\/wp\/v2\/categories?post=927837"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.europesays.com\/us\/wp-json\/wp\/v2\/tags?post=927837"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}