They fail because the science was never made legible. We often use the term “early” or the phrase “ahead of its time” to respectfully explain the demise of such ideas. It is a generous epitaph. It is also, frequently, an inaccurate one.

This is the quiet problem in the room, sitting there like an unclaimed suitcase at an Indian railway station. Everyone can see it. No one wants to touch it. To name it feels impolite, as though one is criticizing the founder, the invention, the years of stubborn labour, the lonely conviction that dragged an idea from paper to prototype. But this is not a criticism of the scientist or the entrepreneur. It is a criticism of the passage between invention and adoption.

That passage is badly designed. And the problem has a name: the translation gap.

Deep-tech founders often believe that once the technology works, the world will eventually understand why it matters. It is a noble belief, and like many noble beliefs, it is only partially true. The world does not adopt technology in the order in which it is invented. A laboratory asks: does it work? A market asks: can I trust it, use it, pay for it, procure it, maintain it, regulate it, and explain this decision to someone else without looking like a fool?

The market does not compare technologies the way engineers or scientists do. Markets compare risk. And the company that makes its risk most legible often gets the first chance to prove itself.

TRL is not enough

Deep tech already has a language for technical progress: Technology Readiness Level, or TRL. It is a useful scale. It tells us whether a technology has moved from principle to proof, from proof to prototype, from prototype to demonstration in a relevant or operational environment. This matters greatly. A technology that does not work cannot be rescued by poetry, narrative architecture, or a founder speaking solemnly about transformation in a room full of people who cannot yet explain what the company does.

A company can be at TRL 7 - a working prototype demonstrated in the field and still be completely unreadable to the people it must persuade: investors, buyers, regulators, partners, users, procurement teams, hospital administrators, factory managers, defence evaluators, and so on.

The technology may be ready. The company may not be. Or more precisely: the company may not yet have designed the path by which the world can understand, trust, adopt, and defend what has been built.

This requires a second framework. Call it World Readiness Level - WRL.

WRL is not a measure of scientific maturity. It is a measure of legibility. It is proposed specifically as an internal framework for deep tech teams, to sit alongside TRL rather than replace it.

TRL asks: does the technology work? WRL asks: can the world make sense of it?

Can the buyer name the problem in their own words? Can the first pilot prove a commercial assumption, not merely a technical one? Can an investor explain the company without the founder in the room? Can the first user encounter create trust instead of confusion? Can the company name its risks honestly, sequence them intelligently, and show how each milestone reduces them?

A company with low TRL is not ready. But a company with high TRL and low WRL is trapped in a more painful condition: it has solved the hard scientific problem and left the equally hard human problem untouched. That is where many promising companies get stuck - not at invention, but in the passage from proof to adoption.

The prototype is not the product

One of the most expensive misunderstandings in deep tech is the belief that a prototype is nearly a product. It is not.

Between these two states lies the territory where many technologies disappear. Not dramatically. Not because the science was incomplete. Not because the founder lacked courage. They disappear slowly, respectably, almost bureaucratically into endless pilots, custom deployments, technical tutorials, unclear procurement conversations, and investor meetings that begin with excitement and end with everyone needing a bit more time to understand the space.

The technology survives, but the company does not fully form around it.

Consider a company that builds a sensor outperforming available alternatives under harsh field conditions. The data is cleaner. The hardware is robust. The pilot technically succeeds. Everyone nods. Slides are made. Photographs are taken beside industrial equipment. Then nothing happens. The buyer does not renew because the dashboard does not map to an existing decision, no clear budget owner was identified, maintenance responsibility was never resolved, and the output created more organizational work than the problem it claimed to solve.

The sensor worked. The adoption failed.

Or consider a med-tech company with a diagnostic tool that answers a real clinical question with genuine precision. The trial data is promising. The science is sound. But inside the hospital, nobody can agree whose workflow it changes, which department pays, what the first deployment proves, or how the result alters a clinical decision. The machine may be brilliant, but the institution remains unconvinced.

The problem is not whether the technology works. The problem is whether the world around it has been designed to receive it.

Translation Design

The discipline that closes the gap between TRL and WRL does not yet have a proper home. It borrows from product design, systems thinking, service design, strategic communication, venture building, and industrial deployment. But it belongs fully to none of them.

Call it Translation Design - or design for deep tech.

This is not about making deep tech look better. It is about making deep tech easier to trust.

Translation Design is not visual design, branding, deck-making, or the application of a tasteful colour palette to a confused company. At its core, it is the set of decisions that determines how a technology becomes understandable, adoptable, and trusted. It shapes the first use case, defines the first proof, decides what the first pilot must establish, and clarifies who the first buyer is and what risk they are actually taking.

It answers questions the technology alone cannot answer: What is the company asking the market to believe? What must be proven now, and what can wait? Who encounters the product first, and what must that encounter establish? What does the pilot prove besides the fact that the system functions? What would make a serious buyer say- not “this is interesting” but “this is necessary”?

These are not good-to-have questions. They are company-building questions. They shape capital, pilots, regulation, manufacturing, positioning, and the company’s ability to withstand a serious competitor when one arrives.

WRL is not Go-to-Market

It is tempting to say this is simply go-to-market strategy. It is not.

GTM asks how a product reaches a market. WRL asks whether the market can yet make sense of the product. That distinction matters enormously in deep tech. A company can have sales conversations before it has adoption readiness. It can have pilots before it has proof. It can have investor curiosity before it has conviction. It can have an impressive demonstration and still leave the buyer unsure what decision is being requested.

This is why so many deep-tech companies mistake activity for progress. They have meetings but not momentum. Pilots but not conversion. Technical validation but not institutional confidence.

WRL makes these gaps visible early. A high-WRL company can answer these questions with unusual clarity: What specific risk has the technology already reduced? What risk remains, and who bears it? What must the first deployment prove commercially and institutionally, not just technically? Who must trust the product before adoption can happen at scale?

Legibility is not the same as simplicity. A legible deep-tech company does not pretend the science is easy. It makes the shape of the company understandable without flattening the science. It names the problem precisely. It defines the first wedge honestly. It explains the moat without drowning the listener in features. It makes clear what has been proven, what remains uncertain, and what the next stage of capital or deployment will establish.

The actual cost of illegibility

Illegibility is not a cosmetic weakness. It is an operating cost.

First, it costs time. Investor meetings become tutorials. Customer conversations become lectures. Pilots are delayed because nobody knows exactly what is being tested. Partners cannot place the company. Talent struggles to understand the mission.

Then it costs money. A company with weak legibility raises later, raises less, or raises from investors who do not fully understand the risk they have taken. It may lose to a competitor with inferior technology but a clearer product, a sharper narrative, and an easier adoption path.

This feels unjust to technical founders. Sometimes it is. But often it is not a market failure. It is a design failure.

A buyer is not asking only: is this better? The buyer is asking: will this work in my environment, with my team, under my constraints, without creating problems I do not currently have? An investor is not asking only: is this technically impressive? The investor is asking: can this become a company, or will it remain an invention? A regulator is asking: can this be trusted under failure, misuse, variation, and scale? A first user is asking: what does this ask of me?

Every one of these is a design question.

Progress is sometimes reduction

Deep-tech founders are often encouraged to expand the story. The market is large. The platform has many applications. The science can transform multiple industries.

Sometimes this is true. At the early stage, it is almost never useful.

A broad story can create excitement but rarely creates adoption. Adoption needs a wedge - a first context where the technology is not merely applicable but necessary. A buyer whose pain is sharp enough to tolerate novelty. A proof milestone that changes the quality of belief, not just the quantity of interest.

In the first round of capital, conviction and depth may carry the company. In the second, the company must show learning. What did the pilot teach? Which assumption changed? What became sharper? What did the team decide not to build? Which market was postponed because the adoption path was too slow?

Progress in deep tech is not always expansion. Sometimes progress is reduction.

A company that returns from the field with a narrower, harder, more precisely defined thesis has often made real progress. It has translated experience into strategy. It has learned not only what the technology can do, but where the world is most ready to receive it.

That is a company worth backing again.

Deep Tech is not SaaS in a lab coat

Software is forgiving in a way that deep tech is not.

A SaaS product can be released imperfect and corrected through use. Users shape it. Behaviour data reshapes it further. The product mutates toward product-market fit in real time - numbers move, cohorts reveal, the roadmap adjusts. This is not a weakness. It is the structural advantage of software: adoption and iteration happen simultaneously, and the market is always teaching the product what to become.

Deep tech does not have this luxury. It is binary at every meaningful threshold.

The material either behaves under field conditions or it does not. The regulatory body either approves or it does not. The hospital either deploys or it does not. The supply chain either scales or it does not. There is no gradual drift toward fit, no user cohort slowly pulling the product into shape. Each gate is a cliff edge - cleared or not cleared. And unlike software, the cost of returning to the previous gate is not a sprint. It is often months, capital, and credibility.

This is why forcing software-shaped expectations onto deep tech is not merely unhelpful. It actively misleads. Speed metrics, conversion funnels, growth curves, shortened cycles - these assume a product that improves by exposure. Deep tech improves by proof. The two are not the same rhythm, and managing one with the instruments of the other produces decisions that look rational and are actually dangerous.

The prototype must become a product. The product must become a system of trust. The system of trust must become a company. None of these steps can be iterated into existence. Each one must be designed, proven, and earned.

Narrative is not hyperbole

Many technical founders resist narrative because they associate it with exaggeration. They have seen thin ideas dressed in large language. They have seen pitch decks confuse ambition with evidence. They have seen mediocrity described as inevitability. That suspicion is understandable.But the answer to bad narrative is not no narrative. It is disciplined clarity.

Communication is not decoration around the truth. It is how the truth travels.

This matters especially in India, where the ecosystem still too often waits for external interpretation: a foreign pilot, a US customer, a global accelerator, a strategic investor from abroad. These signals can be useful. But when domestic conviction depends on foreign validation, India outsources the interpretation of its own technology.

That delay has consequences. It slows capital. It weakens terms. It trains founders to seek permission from institutions that may not understand the Indian use case as deeply as the Indian market itself does or worse, move base.

The fix is not for founders alone. Domestic capital must sharpen its reading of technical risk, commercialization timelines, pilot quality, IP strength, manufacturing feasibility, and founder learning velocity. Institutions must build better test-beds, procurement pathways, and university-industry bridges with commercial teeth.

But founders cannot wait for the ecosystem to become enlightened.

The companies that become legible earliest to domestic investors, domestic buyers, domestic talent, and domestic institutions will not need to wait for the outside world to grant permission.

This is where World Readiness Level matters.

A deep-tech company that builds WRL alongside TRL behaves differently from the beginning. It designs the first user encounter before the pilot is scoped. It defines what the pilot must prove commercially and institutionally, not just technically. It writes the commercial narrative before the investor deck is assembled. It names its risks before the investor has to discover them. It chooses a first market not because the TAM slide is large, but because the adoption path is credible.

It also understands that every proof point has an audience.

A technical milestone proves something to the team. A pilot proves something to the buyer. A regulatory milestone proves something to the institution. A manufacturing milestone proves something to the supply chain. A renewal proves something to the market.

A strong deep-tech company knows which proof is being created for whom.

That is WRL.

Not softness around the science. Not storytelling as theatre. Not polish after the fact. WRL is the discipline that allows the science to survive contact with the world.

A company earns trust in stages: first from the lab, then from the first user, then from the first buyer, then from the first investor, then from the first regulator, partner, manufacturer, and market.

TRL earns trust from the lab.WRL earns trust from the world.We will need both.

I got my hands on the deliciously inconvenient work of economist Mariana Mazzucato, The Entrepreneurial State. A book that does to capitalism’s favourite bedtime story what a well-hit ball does to a glass window.It shatters the notion that innovation is the exclusive province of bold private entrepreneurs, while the state is at best a clumsy administrator and at worst an active obstacle.

Her argument is simple, devastating, and supported by a heap of evidence. The state is not merely a sleepy regulator wandering in after the exciting people have left the room. Nor is it just a mechanic called in to fix “market failures.” At its best, the state is an investor, a builder, a risk-taker, and a creator of markets that do not yet exist.

Private capital, with honourable exceptions, rarely wanders cheerfully into the swamp of uncertainty where basic science lives. It prefers roads, not jungles. It likes a spreadsheet, a horizon, a discount rate, a visible exit. That is not wickedness, it is structure. Venture capital and corporate finance are not designed to spend twenty years wondering whether a difficult scientific bet may someday become commercially meaningful. They are designed to arrive once the fog has lifted, the map has been drawn, and the larger reptiles have been dealt with.

She points out that Apple did not invent the internet, GPS, the touchscreen, or the lithium-ion battery. It did something else, something highly valuable, elegant, and immensely profitable. It assembled these things beautifully, polished them until they gleamed, and sold them to the world. But the deeper risk, the original uncertainty, the long years of failure and patient public expenditure, lay upstream. That frontier had already been cleared.

The internet, GPS, semiconductor advances, mRNA platforms, space launch systems, advanced batteries, quantum research, and much of modern computing are similar endeavours in spirit, coming from ecosystems in which states funded uncertain science, public institutions built talent and infrastructure, and private firms later commercialised, scaled, packaged, and captured value.

Mazzucato’s point is not that business is useless. That would be childish. Her point is that business often innovates downstream, while the state shoulders the more thankless work upstream. The problem arises when risks are socialised but rewards are almost entirely privatised, when the taxpayer finances the difficult years, but the upside is captured later by firms that arrive once the probabilities look friendlier. Such a system is not merely inequitable. It is strategically warped.

Now here is where the story becomes especially interesting if we use Mazzucato’s lens for India, not as abstraction, but as history unfolding in sequence.

At Independence

India inherited not a modern innovation economy but a colonially distorted one. Scientific capacity was thin, industrial capability weak, and research institutions few. There was no sensible reason to suppose that private capital - scarce, cautious, and itself emerging from a narrow commercial base - would undertake the giant civilisational task of building a modern scientific nation.

So the state did it.

In the 1940s and 1950s, it expanded the Council of Scientific and Industrial Research into a network of national laboratories. In the 1950s and 60s, it built the IITs - not as finishing schools for global labour markets, but as engines of domestic capability. It placed bold bets on atomic energy under Homi Bhabha, on heavy industry, and on scientific statecraft as a pillar of sovereignty.

By the 1960s and 70s, this impulse extended outward into agriculture with the Green Revolution, into space with INCOSPAR and later ISRO, and into defence research with DRDO. In the 1970s and 80s, it deepened with the White Revolution and the expansion of cooperative and institutional capacity across sectors. These were not incidental acts of administration. They were foundational wagers on capability. Other significant efforts were setting up steel mills (SAIL), refineries (IOC), and heavy machinery (BHEL) capabilities, which took their time to earn the Navratan and Maharatan titles.

There was a certain audacity in all this. A poor postcolonial state, with immense developmental burdens and far more urgent claimants on scarce money, chose to invest in high science, technical education, research infrastructure, and strategic technological sovereignty. There is something rather magnificent about that. It was not always efficient, and it was not always elegant, but it was ambitious in the grand old way.

License Raj

By the 1970s and 80s, India had built laboratories that could think, but an economy that could not always act. The licence-permit-quota regime, with its intricate choreography of approvals and constraints, often rewarded compliance over creativity. Private firms became adept at navigating policy, but less compelled to invest in technological risk. The state was building capability while, in parts of the economy, dulling the incentives that might have allowed that capability to flourish more broadly.

Liberalisation

The reforms of 1991 did not merely change policy, they changed temperament. Competition increased, licensing loosened, capital moved more freely, and a generation of firms found room to grow. The atmosphere shifted from guarded scarcity to cautious confidence.But liberalisation did not produce, at least not broadly, a great wave of private commitment to long-horizon, research-led innovation.What it did produce was a remarkable and globally significant set of stories.

Indian IT firms - TCS, Infosys, Wipro, HCL - built world-class organisations. They exported competence at scale, created millions of jobs, and placed India firmly on the global economic map. Before India became the world’s back office, there had even been quieter, more uncertain attempts to build indigenous computing capability - HCL in its early years assembling machines in a landscape where global giants had exited.

Alongside this, another, less celebrated story unfolded in pharmaceuticals. The 1970 Patents Act, by allowing process rather than product patents, created the conditions for Indian firms to master reverse engineering and cost innovation. Companies like Cipla, Dr. Reddy’s, Ranbaxy, and Sun Pharma built deep capabilities in chemistry and manufacturing, eventually turning India into the pharmacy of the developing world. If Indian IT exported code, Indian pharma exported chemistry - both resting, in different ways, on foundations shaped by state policy and public investment.

There were other islands of engineering ambition - Tata, Mahindra, Larsen & Toubro, Bharat Forge - firms that built real industrial depth in automobiles, infrastructure, precision manufacturing, and defence. And yet, these remained islands that could have been an industrial continent. They did not coalesce into a broad-based corporate commitment to long-horizon, research-led innovation.

The contrast becomes sharper when one considers moments like the creation of C-DAC in the late 1980s. Denied access to advanced supercomputers, India did what it occasionally does at its best - it built its own. The PARAM series was not just a technological achievement, it was a statement of intent, a demonstration that constraint could be converted into capability through sustained public effort.

Through the 2000s and 2010s, Indian firms grew larger, more global, more profitable. Yet, in aggregate, the reinvestment of those profits into frontier research remained limited. Success was real, but it was often built on executing brilliantly within existing technological paradigms rather than pushing those paradigms outward. The distinction matters. There is a difference between succeeding brilliantly on top of an existing global technology stack and funding the creation of the stack itself. One is commercial excellence. The other is civilisational ambition.

But Why?

It is tempting, at this point, to lapse into moral judgement - to accuse India Inc. of timidity, short-term thinking, or a chronic lack of ambition. Tempting, but not particularly useful.

Corporate behaviour, like all behaviour, is shaped by incentives. And for much of India’s post-independence history, those incentives pointed in directions quite different from frontier innovation.

The long shadow of the Licence Raj: For nearly four decades, success in Indian industry depended less on technological superiority and more on administrative navigation. The decisive skill was not building a better product, but securing the right approvals, quotas, and protections. Firms adapted, as rational actors do. They optimised for the system they inhabited. When liberalisation arrived in 1991, it altered the rules of the game - but habits, like old bureaucracies, do not dissolve overnight. In sectors like IT, the old instinct for arbitrage did not disappear; it merely evolved - from regulatory arbitrage to labour-cost arbitrage. Efficient, profitable, globally competitive - but not necessarily innovation-led.

The nature of the Indian market itself: For much of the last half-century, India has been a country where price sensitivity sits at the heart of demand. When hundreds of millions of consumers are making finely balanced trade-offs at the margin, the market rewards affordability, reliability, and scale far more than frontier experimentation. In such an environment, jugaad - that ingenious, frugal, improvisational engineering we celebrate - was not a cultural quirk. It was an economically rational response. Deep-tech R&D, with its long gestation and uncertain payoff, rarely finds natural demand in a market optimised for cost.

The quiet effect of import substitution: For decades, many Indian firms operated in markets where competition was structurally limited. If you are the only licensed producer, or one of a protected few, the urgency to innovate diminishes. Innovation, after all, is not an act of virtue; it is usually an act of necessity. Where survival does not depend on technological advancement, investment in R&D becomes optional rather than existential.

The peculiar history of intellectual property: India’s 1970 Patents Act, which excluded product patents in pharmaceuticals and chemicals, was a deliberate and, in many ways, wise developmental choice. It enabled the rise of a powerful generic drug industry and made medicines affordable at scale. But it also meant that, for a long period, the incentive to invest in proprietary, high-risk innovation was structurally weaker. One cannot build a culture of IP creation in a system designed, quite sensibly at the time, to prioritise diffusion over exclusivity.

The discipline and limitation of capital markets: Indian capital markets, like most others, reward visible returns. Dividends, buybacks, steady earnings - these are legible signals of performance. Long-horizon R&D, by contrast, is opaque, uncertain, and often thankless in the short term. The widely cited imbalance - where leading firms have returned vast sums to shareholders while investing comparatively modestly in long-term assets - is not merely the choice of corporate imagination. It is a function of what markets have chosen to reward.

None of this is an excuse. It is, rather, an explanation.

Because the crucial point is this: the structures that once made such behaviour rational are no longer as binding as they once were. Competition has intensified. Global markets have opened. Technology cycles have accelerated. The cost of not innovating is rising steadily and geo-politics, above all, is non-deterministic because the rules are now dynamic.

The playbook that built many of India’s most successful companies - protection, arbitrage, efficiency, etc. - was well-suited to its time. It will not, by itself, carry them into the next one.

Then and Now

Curiously, we find ourselves in a familiar phase - one that echoes the early decades after independence, though in a different idiom.

The Indian state is pushing to build, again.

In the digital realm, it has created public infrastructure of a kind few countries have attempted at scale - identity through Aadhaar, payments through UPI, data through Digi-Locker, commerce through ONDC. These are not mere administrative conveniences, they are foundational rails on which private innovation can ride.

For deep tech and frontier science innovation, the ₹1 lakh crore RDI Scheme marks one of the largest single commitments to research and innovation in Indian history. PLI schemes across 14 sectors totalling ₹1.97 lakh crore are explicitly designed to attract corporate investment as co-funders, not just beneficiaries.

ISRO opened the space sector to private companies; IN-SPACe was established (2020) as the regulatory and facilitation body, and the Indian Space Policy 2023 allows private companies to own, operate, and commercialise launches.

The Anusandhan National Research Foundation (ANRF) was established (2023), modelled explicitly on the U.S. NSF, with ₹50,000 crore over five years and a mandate to fund industry-linked research.

The Semiconductor policy deployed a $10B incentive package to attract chip fabrication, with the first fabs now announced and under construction.

The National Quantum Mission (2023) committed ₹6,003 crore over eight years for quantum computing, communication, and sensing.

The Draft National Deep Tech Startup Policy (NDTSP 2023) and subsequent DPIIT/Startup India revisions now extend deep-tech startup recognition to 20 years from incorporation - a direct acknowledgement of the valley of death problem.

Defence procurement policy was reformed, with 75% of the capital budget reserved for Indian industry, while DRDO actively spins out technologies for private sector commercialisation.

Once again, the state is laying track through uncertain terrain.

And here the global evidence becomes impossible to ignore. Across sectors - semiconductors, biotech, space, energy, computing - the pattern repeats with almost tedious consistency. States fund early science alongside venture capital. Public institutions build talent. Governments or large buyers act as first customers. Corporations and startups then scale, refine, and capture value.

The real story is this: almost no foundational deep tech innovation has ever reached global scale without deliberate, structured support from a state, a corporate partner, or both. The question is never whether support happened. The question is only what form it took, and at what stage it arrived.

What emerges from the above data is not merely a catalogue of innovation, but a deeper pattern in the way nations and institutions bring the future into being: that the most consequential technologies are seldom born in the open marketplace, but are instead nurtured within carefully constructed systems that align capital, demand, and purpose over long horizons. In each case, whether in semiconductors, space, energy, or biotechnology, we observe not the triumph of isolated ingenuity and venture capital, but the quiet orchestration of support where the state, the corporation, and the first committed customer together assume the burden of uncertainty, creating in effect a provisional market before a real one can exist. It is this deliberate bridging of the non-commercial phase, this willingness to underwrite risk in pursuit of capability, that transforms fragile ideas into enduring industries; and it is here that the true lesson lies—that innovation, in its highest form, is less an act of discovery than an exercise in collective resolve.

What India Inc. should (must) do

It’s a question that must be answered - what exactly should India’s largest firms do, in what form, and using which mechanisms, to support deep tech innovation? The answer is not abstract. The global playbook is visible enough. What matters is choosing the right role.

1. Become anchor customers for Indian deep tech: Each of India’s top conglomerates should identify a handful of sector-relevant deep-tech startups and commit to pilot purchases, deployment contracts, or off-take agreements. For hard tech, nothing de-risks capital like a real buyer.

2. Create patient capital with an actual deep-tech mandate: Tata, Reliance, Mahindra, L&T, and the larger Indian pharma groups should each run a properly capitalised, independently managed deep-tech vehicle with a ten-year horizon, and with scientists and engineers involved in investment or acquisition decisions rather than financiers that await numbers in excel sheets.

3. Form joint ventures with spinouts from academic institutes: Large Indian firms should build formal pipelines with academic institutes and their research parks or incubators, and co-create spinout JVs where they bring capital, testing environments, procurement, and distribution, while the startup brings the technical edge.

4. Use off-take agreements to build Indian climate tech: Steel, cement, energy, and refining giants should publish medium-term procurement frameworks for Indian-built clean technologies. Not vague aspiration. Actual volume bands, price triggers, and deployment conditions. In capital-intensive climate tech, that is often the bridge between prototype and plant.

5. Turn the diaspora from a story into a strategy: India’s largest IT and engineering firms should create genuine spin-out programmes - not internal incubators, but mechanisms through which senior technical talent can build independent product companies in India, with the parent as minority investor and first customer.

On our current path, India may well become wealthier, more efficient, and more globally integrated. But it risks remaining dependent on external technological foundations. A more ambitious path exists - one in which Indian firms co-create those foundations, building industries that do not yet fully exist.

The state, for all its exasperating habits, its files and forms and fondness for delay, has done something rather remarkable. It has allocated handsome budgets for emboldening institutions, capabilities, infrastructure and founders-to-be. And now, more than ever, it is inviting to build a fantastic runway for frontier innovation to take flight. It is time for the next generation of founders to surface and succeed in bold ambitions, supported by India Inc.

Yesterday, I was having a random conversation with my teammates that began with the state of AI triggered by the question - should we attend the AI summit? The conversation drifted as good ones sometimes do - toward the pace and type of innovation that shapes the trajectory of human civilization, and has there been much done since the 1970s.The answer was democratization of technology - the mobile phone in the hands of millions of people. To which, I asked, perhaps unhelpfully, “to what effect”? Most of the people moved their thumbs - not to make korean finger hearts but to suggest doom-scrolling.

There is a pattern in technology that we rarely articulate, perhaps because it is inconvenient. When a real constraint disappears, the world changes. When it does not, we build fifty versions of the same thing and compete over polish in the name of removing friction.

Look around. Customer relationship management software is a vast and mature market with hundreds of vendors. Nearly every serious enterprise already uses one. The variations proliferate endlessly, ranging from AI-enhanced dashboards, predictive scoring, vertical-specific modules and now moving to auto-capturing conversations, etc. The core problem of digitizing and tracking customer relationships was solved long ago. What remains is refinement.

Note-taking applications crowd every platform. Each promises a better system of tags, backlinks, summaries, synchronization. Yet the fundamental act of capturing a thought and retrieving it later, has not changed in essence. We are optimizing ergonomics, not expanding cognition.

Email clients repeat the ritual. Billions use consolidated infrastructure, yet new “better inbox” tools appear with seasons’ regularity. They are not misguided. They are competing within a finished architecture.

There are millions of mobile applications in existence. Very few remove a fundamental constraint of human life. Most rearrange what is already possible.This is not cynicism. It is pattern recognition.When a constraint has already been eliminated, innovation fragments. Categories multiply. Competition shifts to surface-level differentiation. We optimize inside a solved box.

Now consider the moments when ceilings truly fell (taking some examples post the seventies).

Lithium-ion batteries removed the energy-density constraint that tethered portable computing to walls. Mobility changed. Smartphones became inevitable. Electric vehicles became plausible at scale.

Public-key cryptography solved the trust problem between strangers in digital space. Without it, there is no secure commerce, no digital banking, no internet economy worthy of the name.

3D printing (Additive manufacturing) altered how we think about production. Instead of carving away material, we began building it layer by layer. Matter itself became programmable.

CRISPR transformed gene editing from blunt intervention into programmable precision. Biology moved from observation toward design.

Reusable rockets shifted launch economics by respecting physics more intelligently. Space ceased to be spectacle and began to resemble infrastructure.

These inventions/developments were not category expansions. They were constraint removals. They did not refine the interface to reduce cognitive overload or friction, they altered the boundary conditions.

When Google launched in the late 1990s, it looked like yet another search engine layered atop an already crowded web - could easily be called a wrapper over the internet. Directories existed. Indexes existed. Search existed. But the web’s true bottleneck was not access. It was relevance at scale.Google reinterpreted hyperlinks as signals of authority. It transformed the internet into a computable graph of trust. It began as an application. It became infrastructure.

Everything rests upon something else. The question is never whether you are building on a platform. The question is whether you are altering the load-bearing beams. This is the lens through which new ideas in AI must be examined.

If a company calls a large language model, wraps it in a pleasing interface, and automates an existing workflow a little faster, it may be useful. It may even be profitable. But it is operating within an existing ceiling. If, however, someone reduces inference costs dramatically, or makes AI systems reliably verifiable in safety-critical domains, or solves the memory bandwidth and energy bottlenecks constraining computation, or rearchitects how intelligence operates at the edge - then a ceiling may be shifting.

The problem in my view is not building with AI, it is building because AI exists.

Platforms are seductive. They invite construction. But construction alone is not progress. Progress occurs when something that was previously impossible becomes possible.

Civilizations do not advance because they improve upon what is already standing. They advance because they confront the constraint that everyone else accepts as permanent.

Energy storage beyond lithium-ion remains constrained by density, cost, and materials. Reliable AI in safety-critical systems remains bounded by verifiability and robustness. Compute infrastructure is limited by energy per operation, memory bandwidth, and supply chain concentration. Materials discovery moves slower than our ambitions require. Water and physical infrastructure challenges persist regardless of how many dashboards analyze them.

These are ceilings.

They are highly scientific/technical, stubborn, often unglamorous. They do not lend themselves to pitch decks that promise effortless scale. They require patience, capital, and a certain tolerance for physics.

If you find yourself entering a category with dozens of nearly identical competitors, pause. The problem may already be solved. You may be refining.If you find yourself confronting a bottleneck that others avoid because it is slow, complex, or governed by hard science, pay attention. That is where leverage hides.

India, of all places, should understand this instinctively. India’s legacy is not merely jugaad, the clever hack assembled with wire and willpower. That is the caricature. The deeper truth is sterner, grander.

It is the baori, the stepwell, where water scarcity was not met with complaint but with geometry, where drought became architecture and necessity descended in spirals of stone. It is the Bhakra–Nangal Dam, where a young republic looked at a volatile river and decided not to fear it but to discipline it despite constraints, converting monsoon fury into electricity. It is the Konkan Railway, carved through mountains that preferred to remain mountains, where engineers persuaded unstable laterite and monsoon-soaked soil to hold steel tracks along a coastline that seems impractical. And it is Mangalyaan, where we borrowed the Earth’s own gravity as a slingshot because our rockets could not afford extravagance, so we used celestial mechanics as thrift.These were not acts of fashionable disruption. They were acts of confrontation.

Innovation, at its finest, is not aesthetic restlessness. It is structural defiance.

We must not build AI for the sake of AI, but build where a constraint dies, a category is created or an industry is transformed.

Let me begin with a small confession.This morning, before the tea had properly steeped and before my spectacles had fully accepted the shape of my face, I watched an AI cheerfully autocomplete a function I had half-typed. It did so with the confidence of a man who has never had to reboot a server at 3 a.m. while muttering apologies to an imaginary customer. And I thought to myself, not for the first time: well then… we’ve clearly crossed a line.

I still remember the sound.

That shrill, wounded-animal scream of a dial-up modem in the late ’90s, negotiating a fragile ceasefire with the internet. You didn’t log in back then. You arrived hesitantly, hoping the connection would hold long enough for Netscape to load and not collapse into a sulk.

Software lived close to consequence in those days. Memory was finite. CPUs were honest. Bugs were personal. When something broke, it was usually because I broke it. There was nobility in that intimacy.

I romanticise those years shamelessly. Deployments by FTP. Shell scripts held together with faith and bad punctuation. The quiet, terrifying magic of a Makefile. ipchains standing guard like Heimdall at the gates of a shared Cobalt RaQ server. I was “full stack” before it became a LinkedIn affectation. The romance was real. So was the fragility.

Those systems collapsed under their own success. Brittle monoliths. Architectures that demanded constant firefighting just to remain upright. Many of our so-called heroics were not virtues, they were symptoms of immature tooling and a complete absence of safety rails.

Zoom out far enough and a simple pattern emerges. Software engineering, stripped of fashion and jargon, has always been an attempt to tame three unruly beasts:

Data. Scale. Complexity.

In the beginning, we fought all three simultaneously and mostly lived to fight another day. Data lived wherever it could fit on RAID arrays chosen after long arguments about read/write ratios. Scale was a prayer. Complexity was managed through caffeine, and on especially bad nights, beer.

Then came Linux, followed by MySQL, PHP, and Apache - later canonised as LAMP. This was a turning point. Databases grew up. We learned that data is not just storage, it is gravity, liability, and memory. Flat files gave way to schemas. Indexes replaced incantations. Constraints replaced hope.

Much later came the cloud, NoSQL, message queues, and their many cousins. Almost overnight, scale was tamed. We stopped asking, “Will this survive traffic?” and started asking, “How much will this cost if it does?” Servers became cattle. Monoliths fractured into microservices. Capacity planning became a spreadsheet conversation. Scale shifted from existential dread to a billing line item.

This wasn’t merely convenience. It rewired how we imagined software. You could assume growth. Recklessness became, if not wise, at least affordable.

And now we arrive at the present time: AI and the promise of addressing complexity. Not computational complexity - that battle remains locked at the gates of mathematics but human complexity. The exhausting translation of intent into implementation. “Tell me what you want, not how,” the machines whisper, confidently.

It is seductive.And it is treacherous.

By 2025, the evidence was already awkwardly clear. AI helped in some places - greenfield projects, well-bounded tasks, less-experienced developers, redesigned workflows. But in mature systems, those with history, scar tissue, and tribal knowledge - it would slow experienced engineers down. Time leaked into prompting, reviewing, and correcting suggestions that looked right but weren’t quite. The machine sounded confident. The code demanded supervision.Velocity hid risk.

Security reports told a less flattering story. A disturbing share of AI-generated code shipped with vulnerabilities, not trivial syntax errors, but deeper architectural and logic flaws that evade casual review. Syntax improved. Judgment did not. Complexity hadn’t vanished, it had mutated, becoming probabilistic, quieter, and harder to reason about.

And yet, here we are in 2026, where AI can stare into the abyss of your 2003-era PHP monolith, whisper a digital “bless your heart,” and in seconds transform spaghetti code into a modular, TypeScript-flavoured masterpiece. It will be clean. It will be DRY. And you would probably be judged for your nested if statements more harshly than a Michelin-starred chef judging a microwave burrito.

This is no longer just autocomplete with opinions.

The newer generation of models like Claude and its peers, paired with tools like Cursor, Devin, and agentic workflows, carry far deeper context windows, stronger reasoning traces, and memory mediated through workspace state rather than a single clever answer. They don’t merely suggest code anymore, they observe systems, traverse repositories, run tests, file pull requests, and wait patiently for feedback.

These agents don’t stare blankly at legacy code. They walk it. They map dependencies, infer invariants, notice patterns that humans stopped seeing years ago out of sheer exhaustion. Given time and guardrails, they can potentially refactor entire subsystems not as isolated snippets, but as coordinated change.

This is real progress.And it matters.

But here’s the catch - understanding is not the same as responsibility.

The agent may grasp structure, but it does not own consequence. It can preserve intent disturbingly well, even when that intent is flawed. Business logic written under outdated assumptions is faithfully migrated, lovingly documented, and efficiently parallelised. The bugs survive not because the AI is careless, but because it is obedient.

This could result in a new class of problem - logic that is wrong at scale, cleanly typed, exhaustively tested, and deployed with confidence.AI no longer produces messy spaghetti.It produces beautifully plated technical debt - sushi-grade, provenance-tracked, and still capable of giving you food poisoning if the fish was off to begin with.

The danger is no longer bad code.It is convincing code.

Which brings us to the uncomfortable question: what does it mean to write code now? The modern programmer is no longer a typist of logic. Syntax is solved. Boilerplate is dead. The job has shifted upward. Today’s engineer is a translator of intent from ambiguity to constraint, from desire to system. Soon enough, programmers will decide what not to build. They will prune early. They will sense where complexity will metastasise and cut before it spreads.

And the mythical 10× programmer? They still exist, but not by default.

Once, 10× meant output. More code, faster, leading and teaching several programmers by them just reading the code. Today, 10× means leverage over data, scale, and complexity. It means using AI as a force multiplier without surrendering judgment. Delegating the boring. Distrusting the plausible. Obsessing over the edges where systems age, fail, or get attacked.

This is not automatic. It will need to be learned. It will emerge and persist where taste, discipline, and deliberate practice meet powerful tools. Without that fluency, teams often move faster and accumulate debt even faster.

The real work now is thinking in time.

How will this age?Who will maintain it?What breaks first and how quietly?

These are not questions machines are particularly bothered by. Humans must remain responsible for them.

So no, coding is not dying.It is shedding skin - yet again.

We have moved from wrestling bits to managing probability as intelligence for raw material. The intoxication is real. The dangers are measurable. The progress is undeniable.

And for the first time, the most dangerous code in the system may be the code everyone agrees looks correct.

And yet, time after time, I have also watched their work stall - not because the idea was flawed or because the science was weak, but because the journey from discovery to deployment in India resembles the 14th century Marco Polo map - long, uneven, and littered with challenges no one can quite predict or explain.

India now awards somewhere between 24,000 and 30,000 PhDs each year, depending on which dataset you pick. That makes us one of the largest producers of doctorates on the planet. But our Gross Expenditure on R&D remains stuck around 0.65 percent of GDP. In 2024, Israel spent close to 5.8 percent, South Korea 4.9, the US 3.5, Japan 3.3, Germany 3.1, and China 2.4. These are not marginal differences. They are philosophical ones.

We produce scholars at scale but fund science at austerity. The result is a reservoir of intellect fed by a very narrow financial stream.Still, India being India, the impossible occasionally happens on its own. Resident inventors now file more than half of all Indian patent applications. DPIIT recognises nearly 1.5 lakh startups, many of them deep-tech. Women lead about 18 percent of these enterprises. Tier-2 and tier-3 founders are no longer outliers, they are common. The spark is not the problem.The channel is.

Most PhD programs still reward novelty over usefulness. A polished manuscript is triumph. A validated prototype is a footnote. We have built a reward system that applauds the first act but never schedules the second. Anyone who has worked in Indian academia knows the rhythm - publish the paper, defend the thesis, bind the document, leave it on a shelf. The last chapter is rarely technology. It is storage.

This is why India needs a dual doctoral pathway.

A discovery doctorate for those who want to extend the frontiers of science. And a translation doctorate for those who want to turn science into tools. Tools that survive dust, humidity, regulation, procurement, and the deeply philosophical complexity of Indian markets.

The translation track would still require peer-reviewed work, but it would also demand a translation milestone - a prototype, a patent with a credible licensee, a public-sector pilot, a regulatory submission, or even a DPIIT-recognised spinout. Coursework would cover verification, TRL/MRL progression, standards, safety, design for manufacturing, quality systems, and ethics, etc. Scholars would go where their work belongs - a district hospital, a power plant, a clean room, a DRDO test range. Not to watch but to learn.

None of this is radical. The UK’s Engineering Doctorate, born in 1992, has scholars spend most of their time inside companies, solving problems that are stubbornly real. Germany has its Industriepromotion, where researchers sit inside Bosch, Siemens and the Fraunhofer universe. Japan quietly embeds its doctoral candidates in the disciplined choreography of regulated manufacturing. These systems treat application not as a compromise but as a craft.

India already owns the scaffolding. ANRF with its multi-year funding mandate. BIRAC for biotech. TDB for scale-up. iDEX for defence innovation. The soil is fertile. The seeds exist. But the irrigation channels are missing. In my reckoning, three gaps must be addressed.

The first is industry demand. Nearly 70 percent of India’s R&D spending comes from government agencies. The private sector contributes modestly. By contrast, private industry funds 60 to 75 percent of R&D in most OECD countries. Expecting companies to absorb translation-track doctoral residents without incentives is wishful thinking. India will need R&D tax credits, matching grants, and lighter procurement pathways to create real demand for such scholars.Think CSR (Corporate social responsibility) and ESG (Environmental, Social, and Governance) incentives as an loose analogues.

The second is intellectual property.This is our great elephant.If a translation scholar builds a prototype under joint guidance from an university, an industry partner, and a regulator, the ownership must be clear. Conversations about academic IP must not head towards unresolved ownership, disputes, and respectful silences. The PUPFIP Bill (Protection and Utilisation of Public Funded Intellectual Property Bill) 2008, was a step in the direction to address this I reckon but it did not go through. A translation doctorate cannot flourish without a national IP template that defines ownership, revenue sharing, licensing timelines, and student inventor rights. Without this clarity, the most promising ideas will be afraid of or die in administrative limbo.

The third is faculty incentives.Indian academia lives and breathes publication metrics. A professor can guide ten PhD students, publish widely and thence climb the prestige ladder. The faculty need reasons to gain professionally from translation outcomes in a similar vein. We need a national rebalancing - patents, licensed technologies, standards contributions, public-sector deployments and TRL verifications must count as scholarly work.

None of this requires a revolution. It requires reordering our incentives and clarifying our structures. It requires acknowledging that a country of 1.4 billion, aspiring to technological sovereignty, cannot rely on accidental translation. We need a system that builds it deliberately.

India’s research landscape is entering a rare moment of alignment. NEP implementation is underway. The Budgets offer funding for bolder steps. Our scholars already carry the ambition, our institutions already carry the mandate. What remains is a doctoral architecture that allows good ideas to complete their journey.

The point of research is not only to understand the world. It is to equip it. India should no longer be satisfied with knowledge stored in archives. The country deserves knowledge that becomes capability. Knowledge that becomes infrastructure. What we imagine in the lab must live in the field. If we redesign our doctoral pathways with this principle in mind, we can ensure that India’s scientific promise does not flicker out at the moment it should finally begin to light the way.

These systems now operate at scales that no human oversight can match. When data drifts, when proxies decay, or when historical bias hardens into policy, these systems may continue to optimize faithfully, not sensing that anything is wrong.Living organisms, by contrast, feel deviation. What we call a gut feeling is the body’s own anomaly detection loop - the sophisticated interplay of interoception, the enteric nervous system, and the insula that senses internal imbalance moments before conscious reason can articulate it.I am fascinated by it and it warrants its own deep-dive post, may be some-time soon.

What if we develop an Ethical-reflex framework that aims to endow AI with an equivalent - an ethical nervous system engineered to register potential harm the instant it appears, rather than waiting for full analytical comprehension.

From an implementation/engineering standpoint though, it might be pragmatic to mimic the body’s immune system (another marvel) where every cell distinguishes self from harm - white blood cells patrol the blood stream, T-cells specialize to fight, and antibodies remember. The immune system doesn’t consult a moral code, it quickly responds. What if algorithms are designed not just to abide by moral codes, but to develop this physiological capability - an internal architecture that senses and corrects ethical drift before harm propagates?

Artificial intelligence was designed in the likeness of the brain. Neural networks mimic perception, reinforcement learning imitates feedback. In the book, Thinking, Fast and Slow, Daniel Kahneman describes the human mind running on System 1 (fast, intuitive) and System 2 (slow, deliberate). While recent research links this dual-process cognition to AI design and using two systems as analogues for fast heuristics and deliberate reflection, with human in the loop most of these efforts remain confined to enhancing cognitive capabilities and to involve/mimic humans in near real time. What is still unchartered, is how machines might self-regulate when those thoughts go wrong - cognition does not equal wisdom.

Ethical Reflex should treat ethics not as cognition but as physiology - a real-time control layer. Where dual-process AI focuses on how systems reason, and it focuses on how they recover, shifting moral reliability from an abstract principle to an operational property. In engineering terms, this turns ethics into a feedback control problem - a matter of latency, sensitivity, and stability rather than intention alone.

To translate this physiology into design, Ethical-reflex would need to function as a parallel, self-correcting process running in the background. This loop mirrors biological homeostasis from stimulus-to-detection-to-response-to-stabilization:

1. Monitor: Continuous collection of model inputs, outputs, and contextual metadata

2. Detect: Statistical tests for divergence, parity drift, etc trigger alerts when distributions deviate beyond tolerance

3. Classify: Detected anomalies are scored by risk as a function of domain sensitivity, affected groups, and legal exposure

4. Respond: Minor drifts self-correct via re-weighting, major ones trigger human oversight, model retraining, or policy review

5. Stabilize: Re-caliberate steady state baseline, if so needed

This approach already exists in fragments - IBM’s AIF360 measures bias, Microsoft’s Responsible AI Toolbox explains errors, etc.Drawing from my little experience in and love for resilience-engineering, we could borrow that playbook. Maybe not to the extent of using chaos engineering yet (someday maybe).

It is a change in philosophy - from working towards ensuring error-free algorithms towards free from impact of errors, to the extent possible - it is a continuum that could turn ethics into an engineering layer of resilience.

Also, fairness is not a constant, it is a climate, shifting with culture, context, consequence and even time. Bias is not only mathematical but sociological. For ethics to matter, it must sense its environment and much like immune systems adapting to local pathogens, ethics’ infrastructure must learn regional priors - demographic distributions, legal constraints, linguistic nuance, and cultural norms. Ethical-reflex must treat fairness constraints as universally shared yet locally tuned. Over time, such systems could build strong understanding of ethical sensitivity shaped by continuous feedback from the environments they run in.

Also, while this may feel counterintuitive in a world that treats data as intellectual property, ethics must exist as open infrastructure. Every verified bias or proxy failure should become a reusable detector, a communal antibody rather than a private bug-fix. When one organization uncovers discrimination in its systems, others should be able to inoculate against it. Such transparency transforms ethics from a proprietary compliance exercise into a shared capability, like a distributed immune system that strengthens the entire ecosystem.

Maybe we don’t need a conscience for machines so much as a compass, a GPS for ethics. Like navigation systems, it wouldn’t dictate destinations, only detect drift. It would sense when a model strays from its ethical lane, recalculate in real time, and guide it back on course. But unlike a GPS, this map couldn’t be static. Fairness changes with context, culture, and consequence. The routes must update continuously, drawn from shared experience - the open, evolving moral fibre based on collective data, learning, over time, where the road itself has moved.

What nature leaves us is not just history. It is a living library of design intelligence, offering endless blueprints for innovation.

What is Biomimicry?

Biomimicry, also called biomimetics, is the practice of consciously emulating nature's forms, processes, and ecosystems to solve complex human problems. As a design philosophy, it views the natural world not as a resource to be extracted or exploited but as an inspiration for building a more sustainable and resilient future.

Janine Benyus, who popularized the field, captured it simply:

“We’re awake now, and the question is: how do we stay awake to the living world? How do we make the act of asking nature’s advice a normal part of inventing?”
— Janine Benyus

She describes nature as Model, Measure, and Mentor.

The practice of biomimicry runs across three distinct levels:

1. Form: This is the emulation of a natural shape or structure for a specific function. A classic example is the streamlined beak of a kingfisher, which inspired the nose design of the Japanese high-speed Shinkansen train. This innovation reduced air resistance and the loud tunnel boom that occurred when the train entered a tunnel at high speeds.

2. Process: This involves copying natural chemical recipes or processes. Novomer Inc., a sustainable materials company, developed a way to create high-performance plastics by converting carbon dioxide, much like plants use it as a feedstock during photosynthesis.

3. Systems: The most complex level, this entails replicating entire ecosystems to create designs that are regenerative and provide ecosystem services. Interface Inc. was attempting to design a manufacturing facility that would function like a forest, with the goal of purifying its own water and storing carbon.

Why Biomimicry matters in Deeptech

The confluence of biological insights and modern technological capabilities could make biomimicry a critical advantage for deep tech innovation.

1. Physics that compounds
   Evolution optimizes under scarcity and constraint- challenges deep tech faces. Copying nature’s free optimizations delivers measurable gains in drag reduction, adhesion, noise control, thermal management, etc.

2. Tools have caught up
   What used to be vague (fly like a bird) is now manufacturable to the specifics owing to generative design, high-fidelity simulation, and additive manufacturing.

3. Markets (could) reward it
   Bio-inspired solutions are often cleaner, quieter, and less energy-intensive -attributes regulators and lenders now prize. This would help adoption and lower barriers to scale.

Nature’s playbook in action

Just a short list of some bio-inspired ideas:

Drag reduction - Shark skin - Dermal denticles reduce drag in air and water. Tested on racing yachts and with potential in aviation.

Impact resistance: Mantis shrimp club - Its structure dissipates shock and resists fracture. Inspiring aerospace composites and lightweight armour.

Noise control: Owl wings - serrated feathers break up airflow, enabling silent flight. Adapted for quieter turbines and fans.

Surface engineering: Lotus leaves - Micro/nano-structured surfaces cause water to bead and roll off. Basis for self-cleaning coatings in textiles and solar panels.

Adhesion: Gecko feet - Microscopic hair exploit van der Waals forces. Advanced adhesives now integrate sensors for intelligent gripping, critical in robotics and medical tapes.

Water purification: Aquaporins - Proteins that regulate water in cells inspire next-gen desalination membranes - faster, more energy efficient purification.

Waste circularity: Fungi - Uses mycelial networks to break down industrial waste into low-carbon building materials, reducing toxicity and emissions.

Carbon capture: Microalgae - 10-15x more efficient at photosynthesis than plants, capturing CO₂ from flue gases while producing biofuels and biopolymers.

Biomimetic Robotics: From rigid to soft

This is where biomimicry gets exciting, as a new generation of machines moves from rigid systems to bio-inspired adaptable ones. The intersection probably began with the work of biophysicist Otto Schmitt, who in the 1950s studied squid nerves to engineer a circuit that replicated biological nerve propagation.

Companies like Festo are building robots that mimic animal locomotion, including the Bionic Kangaroo, which recreates the kangaroo’s ability to recover and release energy with every bounce using pneumatic drives and an elastic spring that mimics a tendon. Their BionicMotion robot is inspired by the elephant’s trunk and octopus tentacles, creating a new class of machines for safe human-robot collaboration in manufacturing environments. In response to a need for more adaptable underwater robotics, new designs are mimicking marine life. Researchers have developed a fully soft RoboNautilus that uses pulsed water jet propulsion like a cephalopod for quiet, continuous swimming with minimal disturbance to marine life. Soft underwater robots that mimic SALP (squid, cuttle fish, etc) inspired locomotion by using an origami-led design that allows it to jet water for propulsion by changing shape. A while back, NASA was working on a concept rover that had squid-like design, with tentacle structures that could harvest power and act as a means of propulsion. The development of RoboBees at Harvard's Wyss Institute, which seeks to replicate the coordinated, swarm-like behaviour of insects, is a clear effort to integrate bio-inspired hardware with bio-inspired software. Robotics is shifting from rigid, dangerous machines to safe, adaptive, and intelligent systems.

Bio-inspired startups driving change

The journey from a biological principle to a commercially viable product is being accelerated by a growing number of startups.

GreenPod Labs: This Indian startup developed active packaging that releases plant-based volatiles to activate the inbuilt defence mechanism of fruits and vegetables, extending their shelf life.

Mycocycle: This company is a prime example of a circular economy in action. It uses lab-cultivated fungi to transform hard-to-recycle industrial waste, such as construction debris and petrochemicals, into valuable low-carbon building materials.

Anew Material: The company creates plant-based solvent and plastic-free coatings and adhesives that emulate the adhesion strategies of mussels, sticky bacteria, and geckos.

Aquaporin: This company has commercialized water filtration membranes that mimic aquaporin proteins found in all living cells.

Not as easy as it sounds.

While the possibilities are immense, the path forward is not without challenges. Translating the intricate complexity of biological systems into scalable, models remains difficult, as simplification can lead to a loss of nuance and effectiveness. This underscores the critical need for new, interdisciplinary collaborations that can bridge the knowledge gap between biology, engineering, and design. It’s also crucial to avoid what Feynman called cargo cult science - which in this context would be superficial copying without true understanding of the underlying principles, systems and side-effects at play.

Biomimicry is not a shortcut around hard engineering but a way to aim for it. The next decade hopefully should see a lot more founding teams who translate nature’s design intelligence into manufacturable, investable, and scalable deep tech solutions.

The next frontier

The future of biomimicry lies at the intersection of biology, computation, and advanced manufacturing. This isn’t just about borrowing forms; it’s about embedding nature’s intelligence into the very systems we build, tackling humanity’s hardest problems: cooling a warming planet, regenerating ecosystems, and extending human health.

Materials with innate intelligence

The next generation of materials won’t be inert slabs of steel or plastic but responsive and adaptive. Advances in 4D printing and polymer chemistry are making it possible to design structures that heal, morph, or respond to their environment.

Think of bridge surfaces that knit together micro-cracks, roads that recover from stress (or monsoon) cycles, or textiles that breathe like pinecones opening and closing with humidity. The same principles would extend to healthcare - implants that recalibrate themselves, prosthetics that adjust continuously to their user’s body, and smart stents that respond to physiological changes. These are not metaphors, they are engineering pathways already in early labs, inspired by biology’s genius for resilience.

Robots that perceive to adapt

The frontier of robotics is shifting from mimicking form to replicating function, especially sensory and neural function. Soft robots modelled on octopus arms and bat wings already exist. The next leap is embodied intelligence - machines that combine flexible bodies with distributed sensing.

A gecko-inspired adhesive integrates a neural-like sensor to grip rough surfaces while knowing when contact is secure. Octopus-inspired robots use suction not just to hold objects but to sense them without a central brain. Layer in swarm intelligence, and you have machines that can self-organize to clean polluted rivers, perform microsurgery, or explore alien oceans.

Regenerative urbanism

At the systems level, biomimicry promises a step-change from sustainability to regeneration. Imagine cities designed to function less like factories and more like ecosystems. Buildings could cool themselves like termite mounds, cutting HVAC loads without a watt of air conditioning. Rooftops could harvest moisture like desert beetles, and forest-like corridors metabolize toxins. Such urban systems would not only minimize harm but actively clean air, filter water, and foster biodiversity.

Cool-down a warming world

Nature has perfected passive cooling strategies. The Saharan silver ant sheds heat in desert extremes thanks to specialized hairs that reflect infrared radiation. The Namib desert beetle harvests water from early morning fog, another form reflects almost all of sunlight.

Translating these abilities into absorption coatings and reflective facades could reshape entire energy markets. Buildings, textiles, and vehicles might be designed to maintain a temperature difference without consuming a several watts of electricity.

Longevity Lessons from Biology

Some species appear to resist ageing altogether. Bowhead whales live over 200 years with robust DNA repair. Naked mole rats show near immunity to cancer. Giant tortoises carry mechanisms for extreme cellular stability.

By viewing human longevity through a biomimetic lens, we may find pathways beyond conventional medicine - regenerative implants that mimic self-repairing tissue, therapies that lend genomic stability, or engineered cells that extend organ health.

From self-healing roads to cities that function like ecosystems, these frontiers share a common thread - a move away from simple imitation, toward deep integration with nature's operating principles.

The shift is subtle yet deep - to not just design things inspired by life but to design life into the things we make.

Brilliant technologies either arrive too early - solutions in search of a problem or urgent market needs go unmet because the science isn’t mature enough. This is not a simple communication gap. It’s a systemic challenge, rooted in the very nature of deep tech - long and uncertain R&D cycles, high capital intensity, and profound technological and market uncertainty.

Navigating this foggy frontier requires more than intuition. It demands a better map.

Over the past decades, different organizations have built frameworks to bring clarity to innovation. Each step in this evolution removes a blind spot, yet none has captured the synchrony between technologies that can ship and markets that can pull.

TRL (Technology Readiness Level): The foundation
First sketched by NASA researcher Stan Sadin in 1974 and codified for the Space Shuttle program in 1989, the TRL scale created a common language for engineers and program managers to manage risk. The nine-level system measures maturity from basic principles (TRL 1) to a flight-proven system (TRL 9). The U.S. Department of Defence (DoD) and the European Space Agency (ESA) adopted it for their own R&D and procurement programs.

MRL (Manufacturing Readiness Level): The build-ability check
The DoD soon discovered that a technology could be flight proven (TRL 9) yet impossible to produce affordably or at scale. This led to the creation of MRLs, a parallel 0-10 scale to assess producibility and manufacturing risk. MRLs ask different questions: Can this be made in a real production environment? Are the materials available? Is the process repeatable and high-quality?

CRI (Commercial Readiness Index): The business case
Venture capital needed a tool to answer the ultimate question: Will anyone buy it? The CRI, created by the Australian Renewable Energy Agency (ARENA), serves as a framework for measuring how commercially mature renewable energy technologies and projects are. Its purpose is to help developers and investors judge whether a given solution is ready to compete in the market and attract sustainable financing.

ARL (Adoption Readiness Level): The systemic view
Even with mature tech (high TRL), scalable manufacturing (high MRL), and a sound business case (high CRI), projects often failed. The U.S. DOE developed ARLs to capture systemic, non-technical barriers. Categories include Value Proposition, Market Acceptance, Resource Maturity, and License to Operate.

Now, in Indian deep-tech conversations, TRL is still the dominant language while MRLs, CRIs, and ARLs barely feature. That leaves us flying half-blind.

The instinct, then, is to combine technical and market axes into a neat 2×2 matrix. But to truly capture deep-tech risk, we need to move from single axes to composite dimensions embedded inside. Here’s a straw-man framework of the 2x2 Deep Tech Readiness Matrix fas a start. It maps solution maturity against market readiness.

• Quadrant 1: Pure R&D. Risk is almost entirely scientific. Endeavours here are funded by grants and university budgets or in some cases, early stage investors. The goal: proof-of-concept.

• Quadrant 2: Strong market pull but immature solutions. The race is against time. Market-shaping strategies like securing anchor customers to help evolve the solution whilst paying are critical.

• Quadrant 3: The tech works, but ecosystem resists adoption due to high switching costs, regulatory hurdles, or flawed business models leading to commercial irrelevance.

• Quadrant 4: The promised land. A mature, producible solution meets a willing and able market. Focus shifts from de-risking to execution and scaling.

Assuming each axis runs 0-10, here are five questions worth asking:

1. Is TDR ≥ 5 and MDR ≥ 3? (will differ for stage of VC, esp. early)

2. Does MDR lag TDR by more than three levels?

3. Could a single regulatory change unlock a leap in MDR?

4. Who is the anchor customer underwriting TDR → TDR+1?

5. What non-dilutive capital (if any) could fund the next quadrant move?

A simplified, pocket-note version, if I may :

This matrix certainly doesn’t capture everything. For instance, event shocks like COVID led to mRNA pole-vaulting from lab to scale in under 12 months. Similarly, cheaper manufacturing often creates demand readiness way faster.

But the exercise matters. A better map means fewer ventures lost in the Valley of Death. I invite folks to help stress-test this model, suggest refinements, and share observations where ventures zig-zagged across the matrix.

In the Prime Minister’s recent address at the World Leaders Forum and earlier, the announcement of developing the Sudershan Chakra by 2035, I reckon the core central message was speed - the idea that for India to lead, it must now evolve at the speed of the developed world(or faster maybe?).India is no longer just imagining deterrence through posture or procurement - but through architecture, through a defence-industrial engine that can move not over decades, but over months and towards missions.

This raises a fascinating question: What is the underlying physics of national speed? What are the core layers of a sovereign defence stack, and how do we generate enduring velocity - the kind that compounds progress and capability over time.

Here’s a lens to view it, a perspective probably not novel but I’m scribbling it out nonetheless.

In an ideal scenario, true strategic autonomy would emerge from mastering and/or owning each layer of the industrial stack - from raw materials to finished platforms. Each layer represents both a current capability gap and a future opportunity for exponential growth. Now, we may not be able to achieve 100% sovereign capabilities across all layers for all platform but it’s an audacious goal worth setting eyes on!

Unpacking the layers/dimensions

Each the layer/dimension represents both a current gap and a future opportunity.

Layer 7 - Platforms: The imperative of speed

At the apex sit the final, finished systems visible to the naked eye: Tejas fighters, BrahMos missiles, Arjun tanks, and the likes and owing to recent changes in warfare, joining this elite group would be drones not Kamikaze but MALE. Here, the opportunity is transformational: collapsing traditional 15-year development cycles into agile 5-7 year sprints through iterative development, continuous user feedback, and even the debatable spiral development approach to dance toe-to-toe with faster than ever changes in technology and warfare.Give upstarts a bigger chance here

I’m sure the model exists, I mean as reference points, Israel's defence industry moves from concept to deployment in the blink of an eye and China's military-industrial complex has accelerated innovation cycles in recent times.

Layer 6 - System Integration: The challenge of complexity

India possesses highly capable system integrators - HAL, BEL, DRDO labs and private players like Kalyani Defence, L&T and Tata Advanced Systems. The unlock here is digital: what if we create a National Defence Digital Twin ecosystem that enables virtual prototyping, system-level simulation, and rapid iteration before physical integration and testing begins.

Imagine creating a unified yet distributed digital sandbox where every component, subsystem, and platform can be modeled, tested, and optimized in virtual environments before the first piece of metal is cut.

Layer 5 - Mission Systems: The software-defined future

Modern platforms are increasingly defined by their software architecture. While India has achieved hardware victories, viz., our AESA radars, electronic warfare systems, and missile guidance packages, the next frontier is cognitive - making these systems think-then-act, faster and smarter.

The opportunity lies in establishing an AI supported, software command capability focused on building open, secure, and modular architectures that form the digital nervous system of future platforms.(Don’t think of Terminator and the rise of machines !)

Layer 4 - Subsystems: The trust equation

Companies like Data Patterns, MTAR Technologies, Astra Microwave, and Centum Electronics are building world-class subsystems that compete globally. The mission now is scaling trust - creating a National Defence Certification framework that establishes Indian subsystems as synonymous with quality and reliability.

This isn't merely about compliance, it's about engineering a national brand of precision and dependability that enables upstarts to get in the game and to unlock global supply chains for exports.

Layer 3 - Components: The ecosystem multiplier

India's semiconductor mission is gaining tremendous momentum, with major fabs approved and construction beginning. But for sovereign capability - we need the full ecosystem - specialty gases, ultra-pure chemicals, precision tooling, and most critically, the skilled talent pipeline that transforms fabs from standalone facilities into innovation hubs.

The multiplier effect occurs when these ecosystems achieve critical mass, creating virtuous cycles of innovation, investment, and expertise development - just like the software paradigm did.

Layer 2 - Advanced Materials: The translation challenge

Indian research institutions produce world-class R&D in advanced materials: silicon carbide for power electronics, high-temperature superalloys for jet engines, composite materials for aerospace applications. The grand challenge is industrial translation: moving from laboratory breakthroughs to factory-scale production.

What if we have a National network of pilot plants - specialized facilities that bridge the valley of death between research and manufacturing, de-risking technologies for venture capital investment and scaling?

Layer 1 - Raw Materials: The foundation of power

India possesses critical mineral reserves - lithium deposits in Jammu & Kashmir, rare earth concentrations in Odisha, titanium sands along our coasts. The ultimate strategic opportunity is value addition - transforming raw geological wealth into refined materials that feed our own industrial stack.

This demands private-led, first-of-a-kind refineries and processing facilities that climb the value chain.

The engineering challenge of our time

From a builder’s lens, one pattern becomes clear if I were to hazard an approach that could help accelerate innovation by startups, the 3 key anchors/levers would be :

1. Pilot infrastructure: The missing middle between brilliant research and dependable, scalable products. India needs a national network of pilot plants, demonstration facilities, and technology transition centres that de-risk innovation and accelerate commercial adoption.

2. Trusting startups to deliver: The responsibility and allowance for innovation must be spread to young companies and founders and for that the traditional L1/L2 model of picking players on the basis of cost-for-specs has to give way to a new paradigm.Both for Sudershan chakra and MALE drones, the approach is to pick L1 and L2 bidders in the ratio of 64% to 36%. this is an opportunity to change - why not carve out 36% for startups and make it an opportunity for them to prove they can deliver.

3. Integrated quality: Crafting a national language of trust where Indian certification becomes a global asset, not just a compliance requirement. This means establishing standards that are not just the baseline for startups to strive for but recognized internationally as well.

4. Software sovereignty: Architecting modular, secure, and upgradeable digital backbone for all future platforms. As warfare becomes increasingly software-defined, this layer becomes the most critical differentiator.

5. Research for capability: Create a Defence Research Grid - a coordinated, mission-aligned network across all leading academic institutions, DRDO, active-duty personnel from the forces and private firms, with accelerated research tracks designed to ignite/guide young minds - a pathway to translate research into capabilities we need.

Warfare is changing, battlefields are moving.The future will not be defined by what we import but it will depend on how quickly we can build what we need next.

India’s defence ambition needs an engine to get there, fast !

Even with this triumph, manufacturing’s share of our GDP remains stuck at around the mid-teens, a figure that has barely budged for a decade.

This raises two critical questions. One, in a world of rising conflict, can geopolitics turn "China+1" into "China+2" just as easily? And two, if our next decade looks like our last, will we end up as just a bigger, more efficient workshop? The real prize isn't in assembling the world's existing products; it's in winning the right to build the world's first.

The First-of-a-kind (FOAK) imperative

A First-of-a-Kind (FOAK) factory is where a laboratory proof is forged into large-scale reality. It’s the closest thing modern industry has to alchemy: the brutal, uncertain process of transforming a scientific blueprint into a physical product for the very first time.

India can’t afford to sit on the sidelines; the next wave of industrial value will be captured by whoever builds the first plants, not the cheapest copies. Meanwhile, policy tail-winds are real - a ₹ 1.97 lakh-crore Production-Linked Incentive (PLI) war-chest now spans 14 sectors. (Press Information Bureau)

Geo politics is real - last I checked, the number of active and armed conflicts in the world was greater than 10 and for an over all view of ongoing conflicts across the planet, check this map.

This isn’t just an opportunity, it’s a national imperative

Picture the following handful of critical possibilities:

• Silicon-carbide wafers that let EV inverters run cooler and 15 % more efficient.

• Sodium-ion and solid-state batteries that shrug off 45 °C summers without thermal run-away.

• Gigawatt-scale green-ammonia units that slash our fertiliser import bill and double as hydrogen export hubs.

• 800 kV HVDC converters to push desert wind from Kutch to distant data-centre with single-digit loss.

• Heat-pump-driven HVAC that halves energy use in offices, malls and metro lines now built for 50 °C ambient.

• Small-modular-reactor (SMR) plants that serial-produce 300 MW-class reactor cores and steam generators.

Someone out there might be thinking far better and different ways to address these - each one is a moonshot until it is not.The folks who crack them at Indian price points, while still earning global-grade margins, won’t just follow the supply chain - they will write it.

Viewed through this lens, its relevance in the Indian context already has bold reference points:

• Biotechnology & Pharma: When a company like Biocon (back in the day) built the insulin plant , that is a FOAK project and what followed were NOAK(nth of a kind)

• Aerospace & Robotics: Agnikul’s 3D-printed engine facility is a quintessential FOAK.

• Nuclear energy: The prototype 500MWe Sodium cooled fast breeder reactor iin Kalpakkam (yet to be commissioned) is a FOAK.

In short, FOAK is the playbook for turning our scientific know-how into industrial show-how across every critical, high-value sector.

FOAK btw is not a new concept, the term slipped into the project finance lexicon in the early 2000s, heavily popularized by the U.S. Department of Energy’s Loan Programs Office (LPO). It was created to give a specific name to the high-risk, capital-intensive chasm between a successful lab prototype and the first commercial-scale factory.

The LPO used the FOAK designation to justify loan guarantees for these pioneering projects, acknowledging a clear market failure where private capital was too risk-averse to fund the crucial first factory. Since then, the term’s relevance has expanded to UK and EU.

In essence, FOAK has become the official signal for the high capital need, learning-curve, certifications and offtake risks that come with building something the world has never seen before.

How does the Western FOAK playbook look like?

In the U.S. and Europe, the deepest tech leaps arrive through First-of-a-Kind (FOAK) demonstration plants—projects too novel for commercial banks, so they lean on bespoke capital stacks, loan-guarantees, and long-horizon offtake.

1. Blended capital – public loan programs absorb downside so private equity can price upside

2. Contracted demand – anchor customers(government?) sign multi-year offtake to de-risk revenue

3. Learning-rate mandates – detailed cost-drop milestones unlock the next tranche of capital.

The question is, can India graft that muscle onto Make in India bones?

The Barbell Strategy: The only path forward

Wouldn’t China+1 be enough?” It’s a tempting thought. It’s lower risk and delivers immediate results.Let’s look at that :

1. The margin race: We compete on being the cheapest and most efficient but and when the optionality argument fades, we could end up competing with Vietnam, Mexico, and eventually countries in Africa maybe. It’s a race to the bottom on margins.

2. No strategic moat: Global brand like Apple or Samsung hold all the power. They own the design, the brand, and the customer relationship. They can decide to move to another country(just as they moved to India),The world has to come to us for that specific technology or process.

3. The ambition ceiling: A pure assembly economy generates jobs, but it rarely generates the kind of high-value, high-wage jobs that create a wealthy nation. You risk getting stuck in the (not-so)middle-income trap, forever executing someone else’s vision as a contract manufacturer.

So, the answer isn’t FOAK instead of China+1 but in addition to. China+1 - a barbell strategy, if you like. We use the “China+1” wave as the safe, steady, cash-flowing side of the barbell. It builds our manufacturing muscle, absorbs labor, and gets us into global supply chains. It pays the bills today.

In parallel, we take high-conviction bets on the other side of the barbell - a portfolio of FOAK projects in critical sectors. This is the high-risk, high-reward play that builds the IP, the margins, and the strategic autonomy for tomorrow.

So no, merely China+1 is good but not enough in the long run - it makes us a bigger factory. The FOAK strategy is what could what make us an indispensable one.

“We live together, act on, and react to, one another; but always and in all circumstances, we are by ourselves. The martyrs go hand in hand into the arena; they are crucified alone. Embraced, the lovers desperately try to fuse their insulated ecstasies into a single self-transcendence; in vain.By its very nature, every embodied spirit is doomed to suffer and enjoy in solitude. Sensations, feelings, insights, fancies - all these are private and except through symbols and at second hand, incommunicable. We can pool information about the experiences but never the experiences themselves. From family to nation, every human group is a society of island universes.”

-Aldous Huxley in The Doors of Perception

When I recently tried the Oculus Quest2 and drowned myself in the available experiences out there, some more than the others heightened the perception in the virtual world - a truckload more than what I experienced moments ago in the real world.Once I took the headset off, the brain and body recalibration took a few seconds or so it seemed. It was fantastic but I don’t think I can do it often.

That is what the metaverse is all about - transporting you into the virtual with a feel beyond the physical. Carl Jung and Aldous Huxley would have loved to experiment in VR and compare to the effect of psychedelics or even use VR after consuming some Mescaline - that would by some hyper experience I wager.

So, now then, virtual reality (VR) heavily relies (read amplifies) upon visual dominance which in turn means that the mind believes what it sees over the other senses and almost all VR experience fabrics are stitched with concepts of neurobiology and human factors.

This means the real measurement is in how disconnected a user is from the real world and thus how immersed and present she is in the virtual. Therefore as a user, I need to maybe wear a few more on body monitors like a smartwatch to know if my adrenalin is not going out of whack to send me into a shock or such. Maybe.

The metaverse has been talked about a lot in the recent month - it’s a digital parallel to the real world. In my reckoning, the pandemic already pushed a lot of us to be more comfortable in the digital more than the physical and which is why many folks are having withdrawal symptoms when asked to return to the physical - work, interactions and the likes. The degree of comfort with which the human race adapted to operate in the digital world in the last almost two years, must have given a lot of confidence to the folks now bent on nudging us to go deeper in the virtual.

If you have seen the movie Surrogates, which was based on a comic by Robert Venditti which in turn was inspired by Cybergypsies written by Indra Sinha. The movie takes it far and beyond where humans sit comfortably in their homes and live in the real world virtually while cyborgs/humanoids or whatever you fancy calling them.

The metaverse is a couple of generations behind that possibility - it’s a fully immersive 3-Dimensional space/environment spawned for you. It’s not bridging the gap between the real and virtual but taking you deeper into the virtual and making it seem real as a result of the spatial experience(s) that the technology can create. Haptic and biometric capabilities will make touch as another experience soon.

You create a version of you to interact with a version of several others in cyberspace and the spatial build-up makes it feel real. Microsoft is recreating the virtual work environment in the same spirit via the Mesh. It’s like captain James T Kirk telling Scotty to beam him up and teleportation holoportation is done!

Loosely analogical to app stores on our smartphones, several communities are driven by creators or the other way round will mushroom in Web3.0 - a version of the internet that warrants separate writing.

VR devices could (would) be the new smartphones. It’s coming, like it or not, this wave where the real world will be the second choice. Homosapiens are on their way to being virtually social and socially virtual beings.

Resignation (late 14th century) seems to be a very negative and inaccurate word as tracing it’s etymology one finds that it stems from resign - to give up/abandon/surrender, which is not what people do, they actually move on to grab what they reckon is better.

Anyways, over a million resignations are expected in the Indian tech industry this year despite over 100% salary hikes in pockets, and attrition is hovering at 20-40% in companies of all sorts regardless of the fact that startups and established technology companies are doling out unprecedented salary hikes for retention and/or new hiring.

Now then, history does not repeat but rhymes, and how?! C, Y2K, Unix, Solaris, SQL, VB, Java, C++, Linux, PHP, mobile-apps, Python, Big data, Hadoop and so on..(keep adding C, Java, and maybe Python at regular intervals there !)

There’s enough memories from my experience of being in tech for more than two decades at least, that the demand-supply imbalance ebbs and flows. Sometimes painfully longer. In fact, don’ t take my word for it and have a look at this old news splinter from 1998.

As for this time, it’s certainly most severe and amplified or so it seems and we shall try and get to the bottom of it, if there is one. But mostly, the problem is chronic and the symptoms are acute. In my opinion anyways and I would divide the problem into 3 parts - Employers, Employees and Environment.

Employers

1. Catching up (almost always)

Most often than not, if a company is a follower and not leader in a tech/tech-applied context then the chances are that they are late to the party. Which means that hiring for experience in a specific skill or set of skills they want to catch-up on, assumes the top-most priority and also, extremely hard. For instance, Big Data, Data lake, Data Science eventually became a mandate for hiring folks who know NLP, CV, AI, ML and so on, depending on the business case, industry, etc.So now, if I did not start capturing user behaviour or such data of relevance, I need to start from all the way back.

Another way to land up in this spot slowly and then suddenly is by ignoring a company emerging from left field which becomes the single largest competitor - easiest example is Tesla for most of the car makers of consequence out there.

This means that unless you have a great culture and inspiring leadership (if that was the case you would be leading the space already), or another company is messing it up on both accounts (eg FB some time back) you would have to cough up a lot of capital to break talent away and on-board.

2. Safety in (inflated) numbers

The migratory season also forces most of the hiring to overcompensate for potential attrition, offer rejections and no-shows based on the most recent trending numbers.Which means, hire 3 for every 1 person leaving, 5 for 2 and so on and in case of fresh out-of-the-college graduates, go 2x or 2.5x of last year’s numbers.

Which means more people who would have been out of the contention post stack ranking, bag an employment offer, which gets interpreted as a higher than actual number of vacancies out there and thus an artificial inflation in demand and also more options to choose from for employees-to-be.

3. Pay more to attract (more)

This is nor merely a ripple effect of point 1 above but also the starting premise - that if you pay higher than the market offers for a role, it offsets the probability of not attracting anyone if you are just a startup, a late comer to the party or a legacy stack company aiming your crossbow at modernisation. And the Beatles kept singing - Can’t buy me love!

Environment

1. It’s all happening here

It’s an year of many firsts - first time so many unicorns, first time so many startup IPOs lined up, first time the Indian startup ecosystem has witnessed the infusion of such insane amounts of venture capital ! And then there are the valuations - the amount that used to be an exit for someone a few years back, is Series A/B today. (Check this)

All of this put together is stupendously fueling the spirit of entrepreneurship - it is almost like a mainstream career choice now, and all the abundant capital with companies there’s immense competition for hiring therefore and precipitating an arms-race like fight for talent(Often called talent before on-boarding and resource in a quarter or two).

2. Remote and very digital

The post-pandemic, remote-work savvy world is ravenously hungry for more digital - another first. The transformations and adoption of technologies has never been faster and more valuable.TCO (total cost of ownership), which was the qualifying criteria for any digital project or innovation idea, got replaced with a simple question - could it help us do more, faster and better? Which means users, customers, businesses and everybody in-between needs to hustle harder than ever.

3. The millennials :-)

Ah, the most spiritual of all human generations in the Anthropocene. They question status-quo, don’t believe anything they are just told, don’t want to sit at one place but experience life like never before and frown at career and life choices of every earlier generation.

They are evaluating their relationship with work ever so more post-pandemic lockdown(s) and by the way, it’s okay for them to just quit or not join an employer or even stop responding to phone calls from the recruiter who had sent an inflated offer letter.

Their needs are probably not captured by the Maslow’s pyramid but maybe a weighted graph of a set of principles or values or both, viz., involvement with freedom, earning with learning, working with purpose, access without ownership and such, where the weights are dynamic! They’re a large chunk of the tech workforce and pretty much shaping the thought process that lead to the onset of this large moving-on.

Some of the other pre-existing challenges are :

4. Quality of engineering education

Dated education curriculum and the resulting employability percentile of passing out engineers is lamentable. From parents, teachers, schools to colleges - the entire system and it’s stakeholders are involved in chicanery to become crème de la crème. Grades, marks, entrance tests, GPA/CGPA, etc - all intended for one sole purpose, qualify (filter) some to be plucked for progress and leave the rest. Now then, at one end we have thousands of students passing the muster of these strong filters and on the other, the vast majority of remnant students that must scamper and join mushrooming private universities to fulfill their dream of becoming an engineer, only to do something outside of the disciple for a living in most cases.(Read this)

5. Incongruous hiring mechanics

Having built and scaled large engineering teams, I can tell you one thing for certain that there is a lot left desirable in the whole chain of events starting from the headcount approval to the on-boarding of a new employee and then the onward journey of this employee until his/her migration elsewhere. Which is why venture investments in HR tech are looking to ride this perfect storm.

It’s a broken telephone from the beginning! Right from the headcount assessment to writing of a job description to filtering of resumes /applications - now done auto-magically by systems called ATS. And then there’s the assessments, interviews and the stream that flows into the on-boarding river.Everyone is playing checks and crosses - the idea in most places is to determine the deficiencies in a candidate and not the potential.But, that’s a Gordian knot for every company/team/hiring manager to untie for themselves.

And it’s a hustle too because there’s a hiring target and timeline for recruiters/agencies to hit that number. Just like sales quota, the unpolished cousin, if you will.

Employees

#Surf’s up!

It is as simple as that for most of the lot, however, they are dragooned into testing what they are worth by some latent triggers like :

1. soon my skill will be baseline or eaten up by this upcoming framework or tech

2. I’m done with my boss/work/company a while back, this is the right time to leave

3. I need to move to a different or better skill/company/location to experience and grow more

From performance appraisals to continuous evaluation, from experience to portfolio, from delivery to outcomes, from specialists to generalists.People want to move on !

There is no silver bullet to abate this great moving-on unfortunately and if anything at all, it would be too little, too late but there’s work to be done to minimise the impact from a repeat in the future.

There have been some serious tectonic shifts in the landscape this time round that warrant addressing the most fundamental aspects for all organisations, big and small, and it is the culture. That’s what disenchants an employee the most.Define a culture and find people who would only come in to solidify the culture and the spirit. Pick a leaf from the story of late Tony Hsieh & Zappos maybe.

And then, there’s the recognition of the fact that tech folks are not harder to hire but looking for a specific skill or experience or both at a point in time when everybody else is doing the same thing, is where the challenge lies.Understand the curves (or trends if you will) of technology - fading, emerging and lasting (like white crew necks, blue denims and canvas shoes!). So, let’s be willing to ditch the fading, early to adapt the emerging and fascinated to cling-on to the lasting.

A good developer must posses at least 2 out of these 3 - skill, domain context and technical acumen.So, would it make sense to hire for acumen, coach and train for skills(ahead in time) and immerse in domain context all along?

PS : But I’m not saying don’t hire for skill or domain context at all !

It all starts with one simple phrase, ‘that’s where the money is’, and everyone turns right!

In case you have not noticed, a lot many are turning towards crypto currencies made famous by the top 3 most talked about and invested in : Bitcoin, Etherium & Dogecoin but now the list runs in thousands.

I would not get into the what is crypto because there’s enough and many people who have explained that but I would like to talk about the differences or the lack thereof between Fiat and Crypto currencies and the dwindling distance between them, or so it seems.

Fiat is government backed and printed “legal tender”. It can be physical or virtual as in electronic/digital - the usage of the latter as we know has increased phenomenally over the last few years.The supply of this currency is controlled by the government and you pay taxes on it and with it.It’s unlimited in supply because all the government has to do is print more.

Cryptocurrency is not backed by any government or bank, and thus is de-centralised and not controlled by a government but an algorithm and you pay taxes on it but not with it.

A cryptocurrency could have fixed/limited or open supply.

While Bitcoin was intended and launched as a peer to peer digital payments, cryptocurrencies are now being recognized and therefore used for transactions on digital platforms that could be used for the physical world as well. Some of the known companies that started to do so are :

1. Microsoft

2. PayPal

3. Square

4. Shopify

5. Twitch

6. KFC

7. T-Mobile

8. NBA

9. Lamborghini(!)

10. Pacaso (Real estate)

Tesla had begun accepting Bitcoins for car purchases but discontinued to do so, thus wiping off a a substantial value from the crypto market.

So the usage of cryptocurrencies as an alternative means to get you by is not theoretical anymore and the question is not if or when but how soon, how many companies and governments will find a way to create intersections of this parallel economy in the making with the regular economy and link to GDP and what not for the average economist to comprehend the real economic state of any country and it’s people.

As of now, the only intersecting point is capital gains tax on selling crypto and converting it to recognised currencies.

Despite the fear of regulations unknown/usage bans and the highly volatile nature of these currencies, several people have taken to this new asset class - over 100 million people in India have pottered about with buying/holding/selling one or more such currencies in the last couple of years alone. As a result, India has the highest number of cryptocurrency owners in the world and most of these people are under 35 years of age.

Riding on this wave of owning and getting rich via crypto trading, is another wave in the making and that is NFT (Non fungible tokens). NFT can be described as (source - Wikipedia):

A unique and non-interchangeable unit of data stored on a digital ledger (blockchain) that can be used to represent easily-reproducible items such as photos, videos, audio, and other types of digital items(files) as unique items (analogous to a certificate of authenticity), and use blockchain technology to establish a verified and public proof of ownership. Copies of the original file are not restricted to the owner of the NFT, and can be copied and shared like any file. The lack of interchangeability (fungibility) distinguishes NFTs from blockchain cryptocurrencies, such as Bitcoin.

Simply put, any form of digital content with unique characteristics can potentially become NFT.Currently, generative art and gaming are the biggest space, followed by digital versions of artists’ (including actors, singers, etc) content.Emails, articles and even law papers are surfacing up as NFTs up for grabs.

A few leaps forward and we bum into the concept of Metaverse (a term coined by Neal Stephenson in his 1992 novel -Snow Crash) which is a future version of the internet constituted by shared (community) virtual spaces in 3D - so everything Internet but constructed in augmented reality. The concept is not new if you have heard of Second Life which was launched back in 2002.

Metaverse is where crypto would play a key role in financial exchanges and transactions as more and more people get lured into the virtual paradigm and need the monies to be able to construct and get by in Web3.0 - which is another name for it, if you’re not confused yet.

So really, cryptocurrency as a tool for financial transaction, economics and capitalism might not even need to cross over in the physical world but eliminate the need for government controlled currencies in the new internet order.

The question is, how far down the road can governments kick this can before they get into the middle of it all by crafting the standards and frameworks that solve for understanding, volatility and the scatter of currencies out there.