Startup and VC Ecosystem Updates | Issue# 17 [September 14, 2026]
What?
In Part I, we walked along the foundations of Indian deeptech – semiconductors, AI, spacetech, electric vehicles (EVs), and drones – the layers where real venture money, and in places real revenue, is already flowing. The rockets are reaching the orbit, the chips are getting designed, and the batteries are getting built. However, that is the half of the story where the outcomes are visible.
This second part turns to the harder-to-engineer half: the emerging frontier. Cleantech and renewables, advanced materials, communications and photonics, biotech, electric aviation, and quantum. These are the sectors where the science is deeper, the capital more patient, and the payoff further out – where a company can spend five years in a lab before it ships anything at all, and where the government, not the venture fund, is often the first believer. The question here is less of “who is winning” and more of “who is quietly building the foundations of the next decade”. Here is more on where that frontier actually stands in 2026, sector by sector.
Cleantech and Renewables: From Megawatts to Molecules
For a decade, Indian cleantech meant one thing – solar panels on rooftops. That notion is now changing. The frontier now is not generating clean electricity but making clean fuels and materials – what the industry loosely calls clean ‘molecules’: biogas, biofuels, green hydrogen, and captured carbon. It is a harder, more capital-heavy game than installing rooftop panels, and it is where the more interesting Indian companies are moving.
The clearest signal is GPS Renewables (~$91.7 Mn raised, ₹635 Cr – ~$66.4 Mn Series C round), which turns agricultural and municipal waste into compressed biogas – a clean fuel made by capturing the gas that organic waste (crop and food scraps, animal and human waste) releases as it breaks down. The shape of its raise matters as much as the size: only ₹125 Cr was straight equity, with ₹200 Cr and an earlier ₹310 Cr arriving not as ordinary equity but as funding tied to specific projects – money committed by a Korean conglomerate and Japan’s Sojitz, alongside Indian Oil to build and own particular plants. This precisely is the fingerprint of a sector that runs on project finance – the money that builds physical plants and rooftop systems – not just venture cheques, because someone has to pay to build the physical plants. GPS Renewables operates joint ventures (shared-ownership projects) with Indian Oil, Bharat Petroleum, and Oil India, and is building what it says will rank among the world’s largest biogas complexes in Kakinada.
Carbon removal is the second frontier, and Varaha (~$34.9 Mn raised, $20 Mn raised of a $45 Mn Series B round) is India’s standout. It works with smallholder farmers to convert crop waste into biochar, then sells the verified carbon-removal credits. Biochar is a charcoal-like material that locks carbon into soil for centuries rather than releasing it through the open burning that is common after harvest. This keeps the carbon out of the atmosphere, where it would otherwise add to global warming. The customer list is the proof that this is real and not a carbon-credit mirage: Google, Microsoft, Lufthansa, Swiss Re, and Capgemini have all signed long-term offtake deals, and WestBridge called Varaha a rare profitable climate-tech company when it led the round – its first climate investment ever.
The hardest frontier is green hydrogen – a clean fuel made by splitting water with renewable electricity, giving off no carbon when burned, and seen as one of the few ways to cut emissions in heavy industry and long-haul transport that batteries cannot easily reach. Newtrace (~$13 Mn raised, $6.3 Mn Pre-Series A round) is the genuine hard-tech bet in this section. It builds electrolysers (machines that split water into hydrogen and oxygen using electricity) with a patent-pending electrode design it claims cuts green-hydrogen production cost by up to 60% – the exact number that decides whether green hydrogen stays a subsidised curiosity or becomes a real fuel. Backed by HDFC Bank and Peak XV’s Surge, and riding the National Green Hydrogen Mission, Newtrace is the sort of deep-materials company that Indian venture has historically shied away from, precisely because the science takes years to prove.
Around these sit the companies focused on deployment and on reuse – putting waste back to work instead of discarding it. SolarSquare (~$100 Mn raised, $53 Mn Series C round) is consumerising rooftop solar – full-stack design, installation, financing, and maintenance for homes – betting on a residential-solar market that, at under 5% penetration against 35% in Australia, has enormous room to run. And Ecoil (~$3.27 Mn raised, $2.5 Mn Series A round) is organizing the scattered supply chain that turns used cooking oil from restaurants and hotels into biodiesel and sustainable aviation fuel (cleaner jet fuel made from waste rather than crude oil) – unglamorous, but exactly the kind of groundwork (gathering the raw material) a biofuel industry needs.
What India is not yet doing at scale is the deep-technology core. Most of the sector remains deployment and services – installing panels, building plants, aggregating feedstock / gathering raw material – rather than inventing the underlying hardware. The electrolysers, the long-duration storage systems (chemistries or the chemical composition of the battery), the carbon-capture equipment: these are where the defensible, high-margin IP lives, and India has only a handful of companies – Newtrace among the clearest – are genuinely building there. It is no accident that the largest raises in this section – GPS Renewables and SolarSquare – went to deployment and project finance, while the hardest science, Newtrace, raised the least.
Thoughts: Cleantech’s structural edge is real – vast domestic demand, a manufacturing base, and a thick policy stack (the Green Hydrogen Mission, biogas mandates, rooftop-solar subsidies) that functions as a standing customer that effectively guarantees demand – the government commits to buying or mandating these clean products, so companies know a market exists before they build. But the sector splits cleanly into two businesses with very different venture logic: (a) deployment and project finance – it can absorb large cheques (e.g., biogas plants and rooftop solar) and generate steady, infrastructure-style returns, but it is capital-heavy and rarely produces the outsized multiples venture investors chase and (b) the deep-technology layer – electrolysers, better battery chemistries for long-duration storage, carbon-capture hardware – is where the defensible IP and the real upside sit, and it is exactly where Indian capital is most hesitant, because the timelines are long and the science can fail. The next few years turn on one question: whether green-hydrogen can be made cheaply enough to compete with fossil fuels, and whether India builds that core technology at home or, as with semiconductors, ends up deploying everyone else’s. The molecule, not the megawatt, is where this sector will be won.
Advanced Materials: The Layer Beneath Everything, and the China Problem
If cleantech is about clean energy, advanced materials are about what that energy – and almost everything else – is physically made from. Battery chemistries, magnets, specialty metals, and coatings are the invisible layer beneath EVs, electronics, wind turbines, and weapons systems. And it is the layer where India is most exposed, because one country dominates it: China accounts for roughly 91% of the world’s rare-earth (a group of seventeen metals essential to magnets, motors, and electronics) refining and about 94% of its sintered permanent magnets (the high-strength magnets inside most motors) and has shown that it is willing to use that grip as leverage, restricting rare-earth and magnet exports through 2025 and 2026. For India, advanced materials is therefore not just a business opportunity but a strategic vulnerability – and a handful of startups are attacking it from several different directions.
The first approach is to remove the dependency altogether. Vimag Labs ($5 Mn Series A round) builds electric motors that use no rare-earth magnets at all, aimed at electric two- and three-wheelers and industrial uses, such as air-conditioning and refrigeration – the price-sensitive, high-volume segments where rare-earth cost and supply swings hurt most. Conventional EV and industrial motors rely on permanent magnets made from rare earths India must import; Vimag replaces them with a “virtual magnet” – software and power electronics that generate the magnetic field electronically, in effect swapping a scarce physical material for code. Vimag now calls itself a scale-up: it plans pilot batches of 1,000 to 10,000 motors by the end of 2026 and mass production in 2027, starting with two- and three-wheelers before moving to cars, buses, and trucks. The claim that its motors match or beat permanent-magnet performance is still the company’s own – it has not yet been independently verified at production scale or locked in a signed carmaker deal.
A second route attacks the chemistry itself. Offgrid Energy Labs ($15 Mn Series A round) is an IIT Kanpur-incubated deeptech company building a battery that sidesteps lithium entirely. Its ZincGel battery technology uses zinc and bromine – both cheap and widely available – with a water-based electrolyte. This eliminates the need for lithium, which China controls. The ZincGel batteries are non-flammable and aimed at long-duration storage. Such storage can hold power for six to twelve hours and the solar energy generated at midday can be used in the evening when demand peaks but the sun has set. Offgrid has built a genuine moat of over twenty-five IP families around them. It is building a demonstration plant in the UK before a planned Indian gigafactory (a very large battery-manufacturing plant), and is already testing with Shell and Tata Power.
Others are trying to make the scarce materials at home. Here, the effort is led less by venture-funded startups than by the government and industrial players: in 2025-26 the Cabinet approved a ₹7,280 Cr scheme to manufacture sintered rare-earth permanent magnets (the high-strength magnets, made from an alloy of neodymium, iron, and boron, that sit inside most EV and wind-turbine motors), targeting 6,000 tonnes of annual capacity. Companies, such as Hyderabad-based Midwest Advanced Materials are among the early movers building this domestic magnet-making capacity – the physical, capital-heavy end of the problem that neither software nor a new chemistry can solve.
A related bet is on high-performance structural materials. Fabheads (~$13 Mn raised, $10 Mn Series A round) designs and manufactures carbon-fibre and composite parts (materials that combine two or more components to be lighter and stronger than either alone) using its own patented 3D-printing technology, for aerospace, defence, shipping, and automotive customers. Aerospace-grade composites are exactly the kind of high-value material India has long imported, which makes domestic manufacturing of them a genuine import-substitution play – and a capital-heavy, slow-building one. A similar argument holds for NoPo Nanotechnologies (~$3.36 Mn raised, $3 Mn Pre-Series A round), which makes high-quality Single-Walled Carbon Nanotubes (an ultra-strong, highly conductive material used in batteries, electronics, and composites) at home.
Some attack the supply problem from the opposite end – recovering critical metals from waste. Metastable Materials (~$2.79 Mn raised, undisclosed Seed round) treats spent lithium-ion batteries not as waste but as ore: it extracts lithium, cobalt, nickel, copper, and aluminium and sells them as refined metals. Its patented process recovers the metals without adding external chemicals, using reactions between the battery’s own materials, and it has built most of its refining machinery in-house. In a country that imports most of its battery metals, turning battery waste back into supply is both a business and a hedge against the very import-dependence this section describes.
And a few sidestep the material question altogether and speed up the search itself. Novyte Materials (₹4.15 Cr / ~$459 K Pre-Seed round), incubated at ICT-NICE, uses generative AI and physics-based simulation to design and test new materials on a computer before they are made in a lab – work it claims can cut research timelines up to tenfold and early physical-testing costs by up to 90%. It is the clearest instance in this sector of the horizontal-AI thread running through this whole series: not a new material, but a faster way to find one.
What India is not doing is the hardest upstream part: refining raw ore into usable, magnet-grade material. India actually holds meaningful rare-earth reserves – the gap is not the rocks in the ground but the capacity to process them. It thus exports raw ore and buys finished magnets back. Even the magnet scheme assumes companies will import much of their processed input, because domestic refining barely exists. This is the thinnest, most capital-hungry, longest-horizon sector in this entire series – materials science demands years of lab work and pilot plants before anything ships, which is precisely the patient, failure-prone profile that Indian venture capital has historically avoided. It is telling that the boldest single commitment here is a government scheme, not a venture round.
Thoughts: Advanced materials is where the “own the defensible layer” thesis meets its hardest test, because the defensible layer here is deep science that takes a decade or more and a laboratory, not a software sprint. India’s structural advantage is a powerful one – a genuine strategic imperative (reducing dependence on China) backed by enormous EV, electronics, and defence demand, which together create a protected home market and a patient government customer. The risk is equally structural: this is the sector least suited to conventional venture economics, where timelines are long, capital needs are heavy, and the science can simply fail. The smartest venture-scale bets are the ones that turn a materials problem into a software or design problem – Vimag replacing magnets with code, Offgrid engineering around lithium – rather than trying to out-mine or out-refine China, which is a sovereign-capital fight, not a startup one. The other routes map onto the same logic: recovering metals from dead batteries (Metastable) or using AI to find new materials faster (Novyte) are asset-light plays a fund can back, while making magnets, composites, and refined ore at home (the government scheme, Fabheads, NoPo) is heavy, patient, sovereign-scale work. The venture money follows the approaches that can be solved with clever software or design; the government is left to fund the physical work – the refineries and factories – that no startup can shortcut.
Over the next few years, the test is whether India can build even a partial domestic supply chain before the next round of export curbs – and whether patient private capital will sit alongside the government long enough for the deep science to pay off. In materials, more than anywhere else in this series, the state will have to lead and the market will have to follow.
Photonics: Moving Data With Light
Every sector so far has been about making or moving something physical – chips, rockets, molecules, and magnets. This one is about moving information, and specifically about doing it with light instead of electricity. As AI data centres run into the same walls the published companion piece described – power, heat, and the sheer volume of data moving between chips – the industry is turning to photonics (using light rather than electrical signals to carry data), because light moves faster, runs cooler, and wastes less energy. The same shift is happening in the sky, where satellites are beginning to talk to each other and to the ground using lasers rather than radio. India has a small but genuine cluster forming around both.
The clearest data-center play is LightSpeed Photonics (~$8.5 Mn raised including grants, $6.5 Mn Pre-Series A round). Inside an AI data centre, most of the data actually moves between chips, and the electrical links / interconnects that carry it slow down, overheat, and waste power under heavy load. LightSpeed builds optical interconnects (light-based connectors that replace those electrical links) in a solderable form / module that slots onto an existing circuit board like any other component – which it claims moves data up to four times faster at about half the power, without the costly chip redesign that rival silicon-photonics approaches demand (where the optical parts are built directly into the chip’s silicon). It is pre-revenue and aiming first at the US, where most of the world’s data centers sit, while planning manufacturing in India.
The larger sub-cluster is optical communication – sending data through open air or space on a beam of light. Velmenni (~$3.67 Mn raised, ~$3.3 Mn Pre-Series A round) works on the ground, building free-space optical and Li-Fi links (data sent as light through the air, rather than through fibre cables or radio spectrum) for the gaps where laying fibre is too slow and radio too congested. It has deployed live, commercial optical links for a private 5G network at a power plant and won a contract to connect Indian submarines in harbour. Astrogate Labs ($1.3 Mn Pre-Series round) builds the laser terminals that let small satellites beam data down far faster than radio can – a real bottleneck, with over 40,000 small satellites expected in orbit by 2030 and radio spectrum already congested. And from the spacetech cluster in Part I, QOSMIC ($3.33 Mn raised in Seed round) builds the optical ground stations that receive those laser signals on Earth. olee.space (~$3.11 Mn raised, $3 Mn Seed round) works on the defence angle – laser-based communication systems – and directed-energy weapons, with quantum-secure protocols on its roadmap. Together, they are the beginnings of an end-to-end Indian optical-communication chain, from satellite to ground.
What India is not doing, once again, is the fabrication. Photonic chips – the specialized silicon that generates and steers light – are made in a handful of foundries (specialized plants that manufacture chips) abroad, the same fabrication gap that shadows the semiconductor story in Part I. India is designing photonic systems and deploying optical links; it is not yet manufacturing the photonic chips underneath them at scale. The cluster is also early and thinly funded – these are Pre-Series-A cheques against a capital-intensive, slow-certifying hardware problem.
Thoughts: Photonics is the sector that most directly rides the two biggest waves in this whole deeptech series – the AI data-center boom on the ground and the satellite explosion overhead – which gives it something rarer in deeptech than clever technology: obvious, growing demand. India’s edge is a familiar one, the same design-and-deploy strength that shows up in chips and space, now pointed at light. The optical-communication sub-cluster is the more investable half today: Velmenni, Astrogate, Qosmic, and Olee are solving concrete, near-term problems (backhaul where fibre can’t reach, satellite downlink, secure defence links) with real deployments and defence demand behind them, which is exactly the kind of protected early market a fund likes to see. The data-centre interconnect bet – LightSpeed – is higher-risk and higher-reward: a genuinely hard problem against well-funded US and Chinese rivals, where being frugal and easy-to-adopt may matter more than being first. The shared ceiling, across both the data-center and satellite plays, is fabrication: as with chips, India can design and deploy photonics long before it can make the photonic silicon itself, and the sector will stay a systems-and-integration play until that changes. But of all the frontier sectors, this is the one where India’s existing strength points at exactly the demand that is growing fastest: the data moving inside AI data centres and between satellites.
Biotech: Engineering Biology
India’s biotech reputation was built on scale and affordability – generic drugs and vaccines made cheaply for the world. The frontier now is different: engineering biology itself. Reprogramming a patient’s immune cells to fight cancer, redesigning microbes to produce materials normally made from petroleum, and using AI to invent drug molecules faster. It is deeper science than generics, with longer timelines and far more patient capital – and a handful of Indian companies are now working at that edge, most of them turning India’s old affordability advantage on the newest science.
The sharpest activity is in cancer, specifically CAR-T cell therapy – a treatment that extracts a patient’s own immune cells, genetically reprogrammes them in a lab to recognize and attack cancer, and infuses them back in. India already has a genuine success here: ImmunoACT (~$15.9 Mn raised, ~$10 Mn Series B round), an IIT Bombay spin-out and SINE incubatee, built NexCAR19, India’s first approved CAR-T therapy – and, rare for deeptech, it is already profitable, posting ₹62 Cr in revenue and a small profit in FY25 while treating over 350 patients at roughly a tenth of international prices. Immuneel Therapeutics (~$49.8 Mn raised, ~$10.5 Mn Series B round) is close behind with its own approved Qartemi CAR-T treatment for certain blood cancers. Pushing the cost frontier further, Cellogen Therapeutics (~$4.69 Mn raised, ~$2 Mn Series A round, plus ₹15 Cr in funding and grants from the government) is developing a bispecific CAR-T, engineered to recognize two of the cancer’s identifying proteins (markers) at once rather than one, so the cancer cannot escape by hiding a single target. With this, the startup aims to bring the cost of a course of CAR-T therapy, which costs $400,000-600,000 in product alone in the West – down to roughly ₹25 Lakh (~$26,000). Both make the same bet – that India’s real edge in frontier medicine is manufacturing world-class therapies at a fraction of cost seen in the western markets, the affordability wedge that has always defined Indian biotech, now pointed at its most advanced corner.
The second frontier is synthetic biology – engineering microbes to manufacture what would otherwise come from petroleum or animals. FermBox Bio (~$14.5 Mn raised, also selected for a ₹25 Cr government RDI backing) uses precision fermentation (programming microorganisms to brew specific proteins, enzymes, and compounds) to make products such as cellulosic enzymes (which break down crop waste into fermentable sugars for biofuel), biofuel inputs, dyes, and cosmetics ingredients, on an asset-light model that leans on contract manufacturing rather than owning heavy plants. Programming here refers to engineering microorganisms genetically and metabolically by altering their DNA and metabolism – to synthesize what we want. Importantly, in 2026, it was among the first startups funded under the government’s ₹1 Tn RDI push to build a large-scale precision-fermentation plant (picked alongside Revelations Biotech, Sea6 Energy, 4baseCare, and Telluris Biotech) for made-in-India cellulosic enzymes. It is the same government-as-co-investor pattern seen across this series (see Part I), now reaching biology.
The third frontier is where biotech meets the AI thread running through this whole series. Peptris Technologies (~$9.17 Mn raised, ~$7.7 Mn Series A round) uses an AI platform to design and narrow drug candidates on a computer before the slow, expensive laboratory testing begins. Its proof point matters: PEPR-124, a candidate for a rare muscle-wasting disease, became India’s first AI-discovered drug to be out-licensed (licensed to another company to develop, in this case, to a US firm) and won an FDA orphan-drug designation (a status that supports drugs for rare diseases). It fits the same pattern the AI blog described – India competing not by spending more than Western labs, but by using software to make a slow, expensive process faster and cheaper. That said, no Peptris-discovered drug has yet cleared a clinical trial, the honest limit on how far this has travelled.
What India is not doing, still, is frontier drug discovery at scale – the decade-long, billion-dollar business of inventing and clinically proving genuinely new molecules end to end. Its biotech strength remains weighted toward services – making and testing drugs for global pharma under contract – and toward making advanced therapies cheaper rather than developing them. Regulatory and clinical-trial infrastructure, though improving (approval timelines have roughly halved in recent years), still lags the US and China, and the deepest-science bets here carry the longest timelines in this entire series.
Thoughts: Biotech is where India’s oldest advantage – doing world-class science cheaply – meets deeptech’s hardest constraint: time. The structural edge is genuine and threefold: deep life-sciences talent, a hospital and manufacturing base that can produce advanced therapies far more cheaply than the West, and a government now willing to co-fund the patient, capital-heavy work through BIRAC and the RDI push. The risk is equally structural: biotech timelines are the longest in this series. A cell therapy or a new molecule can take a decade to reach patients, which sits awkwardly against a venture model that wants an exit inside ten years, and a clinical trial that costs ₹14-15 Cr per product is a heavy lift for an early-stage startup. So the most investable Indian bets are the ones that shorten the loop or lower the cost: affordability plays that take a proven frontier therapy and make it manufacturable and cheap – ImmunoACT has already shown this can work, reaching profitability in its first full commercial year, a rarity anywhere in deeptech – and AI-plus-biology plays like Peptris that turn a slow laboratory problem into a software one, where India’s real moat is often the data rather than the algorithm. Pure frontier discovery – inventing and proving new molecules end to end – will likely stay the state’s to seed for now. ImmunoACT has proven India can manufacture a frontier therapy profitably; the open question is whether India can originate the next one, not just make an existing one cheaper – and whether patient capital, public and private, will wait the decade that takes.
Electric Aviation: The Air Taxi Bet, Grounded in Cargo
The global dream of electric aviation is the flying taxi – small, battery-powered aircraft that take off vertically like a helicopter but fly quietly and cheaply, ferrying people over busy cities. The technical name is eVTOL (electric Vertical Take-Off and Landing). It is a genuinely hard, capital-intensive frontier: the leading Western players (Joby, Archer) have raised billions of dollars over many years, and only China’s EHang has so far won a type certificate to carry passengers. India is a comparative latecomer, but its congested cities and thin road and runway networks make the use case unusually compelling, and a small cluster of startups is now flying real hardware.
The company furthest along toward commercial flight is The ePlane Company (~$21.5 Mn raised, $14 Mn Series B round). Incubated at IIT Madras and backed by the government’s RDI fund, it is now closing a larger round to fund its full-scale aircraft. Its edge is regulatory: it is the first private Indian firm to hold a Design Organisation Approval (DoA) from the DGCA, India’s aviation regulator, which grants official permission to design certifiable aircraft. The startup is also the first to have its own aircraft, the two-seat e200X, accepted into the regulator’s type-certification pipeline (formal approval process an aircraft must pass before it can carry passengers commercially). It has chosen a strong first use case: air ambulances, in partnership with Apollo Hospitals, aiming to cut critical medical transport times sharply, and at the 2026 Farnborough Airshow it unveiled its full-scale PT-01 prototype and signed five supply-chain partnerships toward a made-in-India aircraft, targeting certification flights by 2027 and commercial operations from 2028.
The most prominent pure air-taxi play is Sarla Aviation (~$13.4 Mn raised, $10 Mn Series A round, including a later investment of ~$1.2 Mn by IndiGo Ventures). Named after India’s first woman pilot, Sarla is building the Shunya, a six-passenger air taxi, on a deliberately India-specific bet: carrying more passengers per flight (higher payload) at shorter range, on the logic that Indian trips are short and price sensitivity is high, so payload matters more than range. In 2026, its Sylla demonstrator (a smaller prototype / test aircraft built to prove the design flies before the full aircraft is made) completed over 500 flight tests and became the first aircraft of its weight class built in India to fly. The startup’s backing by IndiGo Ventures, the arm of India’s largest airline, is a demand signal worth more than its dollar size.
The more grounded near-term bet, though, is not people but cargo – and this is where the Indian cluster looks smartest. BLUJ Aero (~$3.38 Mn raised, ~$2.25 Mn Seed round) has built what it calls India’s largest privately developed eVTOL, a heavy-lift cargo aircraft that carries over 200 kg and needs no runway. While its current prototype runs on batteries, the startup is developing hydrogen-fuel-cell versions to extend the range, aiming at logistics, medical supply, and defence supply into places where trucks are slow to reach. Khageshvara Aviation (undisclosed Pre-Seed funding, building tilt-rotor cargo eVTOLs that carry up to 400 kg) is chasing the same heavy-cargo niche for defence, disaster response, and industrial logistics, and offers its aircraft as a service rather than only selling them. In plain terms, tilt-rotor cargo refers to an aircraft whose rotors swivel to lift off vertically like a helicopter, and then tilt forward to fly like a plane.
What India cannot rush is two things: getting these aircraft certified as safe to fly, and building their core parts – motors, batteries, flight-control electronics – at home rather than importing them. No Indian eVTOL has yet carried a paying passenger, and the timelines stretch to 2028 and beyond; the motors, batteries, and flight-control electronics overlap with exactly the imported-components gap that shadows drones and semiconductors in Part I. This is among the longest-horizon, highest-cash-burn sectors in the series, in a global field that is still almost entirely pre-revenue.
Thoughts: Electric aviation is the sector where India’s structural disadvantage (a late start, in comparison to the Western rivals with billions and a long head start) is most neatly offset by a structural advantage the West lacks: cities so congested, and road and runway networks so thin, that the economics of flying over them work sooner here than almost anywhere. The smartest thing about the Indian cluster is that it has mostly refused to copy the Western glamour bet – the intercity passenger air taxi – and gone instead for the use cases with defined, near-term demand: air ambulances (ePlane), heavy cargo into hard-to-reach terrain (BLUJ, Khageshvara), and short high-payload hops suited to Indian trips (Sarla). The risks are real and shared with the whole global field: certification is slow and strict, with regulators demanding exhaustive proof of safety; cash burn is brutal; and the components stack is still imported. But what is worth watching is who India’s winners turn out to be. If they are the cargo and air-ambulance players rather than the air-taxi ones, India will have inverted the global order – proving that the first real market for electric flight is moving goods and saving lives in places the ground network fails, long before it is shuttling commuters over the skyline. That would be a very Indian way to win: unglamorous, demand-led, and built for the country it flies over.
Quantum: The Longest Bet, and the Security Wedge Within It
Quantum is the deepest-science frontier in this whole series, and the one where the gap between hype and useful product is the biggest. It splits into three related fields: quantum computing (machines that use the strange rules of quantum physics to solve certain problems that no ordinary computer can), quantum communication (using those same rules to send messages that are, in principle, impossible to eavesdrop on), and quantum sensing (measuring the physical world with extreme precision). The US and China lead by a distance, pouring in tens of billions. India’s answer is the National Quantum Mission (NQM), a ₹6,000 Cr (~$750 Mn) government programme launched in 2023 that has become the single largest backer of the field here – the government-as-first-investor pattern of this series in its purest form, because there is almost no private market to speak of yet when it comes to quantum.
The clear frontrunner in computing is QpiAI (~$54.1 Mn raised, $32 Mn Series A round that includes backing under the NQM, alongside VC investment). It is building the full stack – the quantum processor, the control electronics, the software – rather than just one piece, and has already put out two working processors: the 25-qubit Indus and the 64-qubit Kaveri. A qubit is the quantum version of the 1s and 0s an ordinary (digital) computer runs on. More and better qubits imply more computing power. Also, qubits vary in error rate, and a better qubit makes fewer mistakes. In August 2026, the startup went a step further and opened what it calls Asia’s largest quantum chip foundry in Bengaluru – its own plant to make the chips, since ordinary semiconductor foundries cannot build quantum chips – targeting a roadmap that runs to thousands of qubits. Notably, QpiAI says it has been profitable at the EBITDA level for three years based on the revenue from its AI-and-quantum software, while the hardware itself matures.
But the more grounded bet – and, we would argue, India’s real opening – is not building the most powerful quantum computer but defending against one. A sufficiently capable quantum machine will one day break the encryption that protects the world’s data today, a looming moment the industry calls “Q-Day.” Guarding against it is a near-term, commercially real problem, and this is where India’s strongest quantum company sits. QNu Labs (~$43.3 Mn raised, ~$21 Mn Series A1 round co-led by the NQM), incubated at IIT Madras, builds quantum-safe cybersecurity, much of which is hardware-based: quantum key distribution (a way of sharing encryption keys that physics itself makes impossible to intercept unnoticed), quantum random-number generators (produce truly unpredictable numbers to build stronger encryption keys), and post-quantum cryptography (new encryption methods designed to survive a quantum attack). It is India’s first company selling commercial, hardware-based quantum security, already deployed across defence, finance, and telecom, with HDFC Bank among its backers and customers. The recent ~$21 Mn raise allows QNu to meet the matching requirement for a ₹150 Cr government RDI project it was awarded – a great example of public and private capital going in side by side.
The same quantum-security logic is being pushed into space. Pramatra Space (undisclosed Pre-Seed round, backed by Pixxel’s founder) is building satellite-based Quantum Key Distribution (QKD) – sending those encryption keys from orbit – and has validated its core photonic chip at IIT Madras, with an in-orbit demonstration planned for 2027. For the uninitiated, satellites can distribute quantum keys across long distances that ground fibre can’t reach (QKD over the ground fibre normally used degrades past a few hundred km). Beneath all of them sits a layer of hardware startups the NQM is deliberately seeding. Among those, Quarkitech (₹2 Cr Pre-Seed round, plus a ₹15 Cr NQM grant) and others are building essential components, such as the cryogenic systems (to keep quantum chips near absolute zero temperature), and specialized lasers and single-photon detectors used to create and read the individual particles of light that carry quantum information, most of which India still imports.
What India is not doing is competing at the frontier of raw quantum computing power, where the US and China are years ahead and spending on a scale India will not match. Nor does it yet make all the deep hardware (the cryogenics, lasers, and specialized components) that its own quantum machines run on at home. This is the longest-horizon, most import-dependent, and most state-reliant sector in the entire two-part series: strip out the NQM’s cheques and there is very little private venture activity left underneath.
Thoughts: Quantum is where the “own the defensible layer, not the frontier” thesis that runs through this whole series reaches its sharpest point. India is not going to build the world’s most powerful quantum computer before the US or China, and it does not need to. Its realistic, valuable win is the quantum-security layer – the tooling that protects everyone else’s data from the quantum machines that are coming – and it is a genuinely good position to hold: QNu is already selling it, banks and defence agencies already need it, and unlike quantum supremacy, it does not require being first in the world, only being trusted and ready before Q-Day. The structural edge is the NQM, which gives India patient capital and a talent pipeline that a pure-venture market is unlikely to sustain for a decade-long science bet. The structural risk is the flip side of the same fact: this is the sector most dependent on the government cheque and least able to stand on private capital alone, and its hardware still leans on the same imported cryogenics and lasers that shadow semiconductors in Part I. QpiAI’s foundry and profitability are a real and surprising bright spot – proof India can build quantum hardware and a business at once. But the honest read is that India’s quantum decade will be defined less by who builds the biggest machine and more by who owns the lock, not the key: the security layer that the entire digital world will have to buy, whoever ends up winning the race to build the computer itself.
The Thread That Ties Them Together
Six sectors, and underneath them the same shape we found in Part I. Across cleantech, advanced materials, photonics, biotech, electric aviation and quantum, India is not trying to out-build the global frontier – not the biggest quantum computer, not the novel major drug, not the cutting-edge photonic fab – and it does not need to. The value it can realistically own sits one layer in from the frontier: clean fuels over raw renewable power in cleantech, recycled and AI-designed materials over mining and refining in advanced materials, the optical link over the photonic chip, the affordable therapy over the invented one, the cargo drone over the air taxi, and the quantum lock over the quantum machine. The same asset-light, defensible, demand-led logic that ran through Part I’s foundations runs through this emerging frontier too – only here the timelines are longer, the science harder, and the capital more patient.
Two threads bind the two parts together. The first is that these layers increasingly feed each other, and feed on the horizontal that sits above them all – AI. It designs new materials (Novyte) and discovers drug candidates (Peptris), and the data-centre boom built to run it is the very demand the photonics layer is racing to serve. Quantum sits slightly apart – its computers may one day accelerate the use of AI, and its security layer defends the data AI depends on. But the pattern holds: Indian deeptech is not a set of silos but a single stack, with intelligence running through most of its floors.
The second thread, louder across this part than in Part I, is the role of the state. In the foundational sectors, the government was emerging as a co-investor. On this frontier, it is often the first investor and the only patient one – the NQM in quantum, BIRAC and the ₹1 Tn RDI fund in biotech and materials, the Green Hydrogen Mission in cleantech, or the RDI backing in electric aviation. This is not our observation alone; by mid-2026, analysts were noting that public capital had moved from backing funds to backing startups directly, taking equity stakes and matching private money rather than only writing grants. Where private venture cannot yet see an exit within its fund’s life, the state is standing in – and the open question, for the next decade, is whether patient private capital shows up to stand beside it, or whether these sectors stay dependent on government money.
That is the honest divide between the two halves of this series. Part I’s foundational sectors are getting funded, and a few are getting finished – chips shipping, rockets in orbit, batteries in vehicles. This frontier is mostly still getting started: the science is real, the companies are early, and the money underneath them is more government than venture. Whether India’s deeptech decade extends from the foundations to the frontier depends less on the technology, which is genuinely world-class in places, than on whether the capital – and the patience – arrives to carry it the distance. The rockets proved India can build hard things. The frontier will prove whether it can afford to.
Thank you for reading through! We genuinely hope you found the content useful. Feel free to reach out to us at ankanatwork@gmail.com and share your feedback and thoughts to help us make it better for you next time.
Acronyms used in the blog that have not been defined earlier: (a) Venture Capital (VC), (b) Artificial Intelligence (AI), (c) Million (Mn), (d) Crore (Cr), (e) Housing Development Finance Corporation (HDFC), (f) Intellectual Property (IP), (g) Indian Institute of Technology (IIT), (h) United Kingdom (UK), (i) Three-Dimensional (3D), (j) Institute of Chemical Technology – Nurturing Innovations and Catalysing Entrepreneurship (ICT-NICE), (k) United States (US), (l) Light Fidelity (Li-Fi), (m) Fifth Generation (5G), (n) Chimeric Antigen Receptor T-cell (CAR-T), (o) Society for Innovation and Entrepreneurship (SINE), (p) Financial Year (FY), (q) Research, Development, and Innovation (RDI), (r) Deoxyribonucleic Acid (DNA), (s) Trillion (Tn), (t) Food and Drug Administration (FDA), (u) Biotechnology Industry Research Assistance Council (BIRAC), (v) Directorate General of Civil Aviation (DGCA), (w) kilogram (kg), and (x) Earnings Before Interest, Taxes, Depreciation, and Amortization (EBITDA).