The Real Risk Is Thinking Small

— a response to Soham’s “Overcoming India’s Technological Cowardice”

There’s a moment in Soham Sankaran’s essay, Overcoming India’s Technological Cowardice, that hit me like a punch in the gut:

We are not poor. We are scared.

He didn’t say that, to be clear. I thought it. But that was my biggest takeaway.

Because for those of us who work in or around Indian innovation, founders, investors, researchers, policy folks, that line isn’t just provocative. It’s familiar. Painfully so.

We’ve all felt it: the meeting where someone says, “Great idea, but India’s not ready.” The pitch deck where a technical risk is marked in red, but a bloated TAM slide gets a nod. The PhD who drops out of academia not to build something, but because they couldn’t figure out who to ask for help.

We are not poor.
We’re just structurally disincentivized to take moonshots.

And Soham lays that bare, surgically, searingly, and with the kind of clarity that makes you sit up straighter.

This isn’t a rebuttal to his essay. It’s a relay baton.
If his piece is a diagnosis of the problem, consider this an attempt to talk about where we go next.

Because there are early-stage investors trying to underwrite technical risk.
There are young founders building category-defining tech in underfunded labs and borrowed space.
And there are regulators who want to support them, but don’t know where to begin.

This post is for all of us who are tired of nodding along in agreement, and are ready to get our hands dirty fixing what’s broken.

First: Soham’s not wrong. But it’s not just cowardice.

The fundamental argument Soham makes is right.
We have been technologically timid. We’ve let entire industries slip away because we didn’t have the collective will, across capital, policy, and talent, to bet on long-term innovation.

But I want to make an uncomfortable addendum.
This isn’t just about cowardice. It’s also about design.

We’ve designed an ecosystem where the odds are stacked against those who dare.

Even when early-stage investors step up (and there are a few of us who do) the drop-off is brutal. There’s often no one to pick up the baton. Growth capital in India is structured to underwrite revenue risk, not research risk. It’s built for pattern-matching, not frontier-busting.

You can’t blame them entirely. Most growth funds were never built to do this. Their LPs didn’t ask for it. Their teams don’t know how to evaluate it. Their fund cycles don’t allow for it.

And so, even the most exciting R&D-first companies, especially in biotech, space, hardware, semiconductors, hit a wall right when they need $15–$30 million to get to the next stage.

That’s the quiet part no one says out loud:
Even if you survive the pre-seed and seed stages in deep tech India, the “Series B desert” is where most great companies go to die.

And when that happens over and over, early-stage capital pulls back too.
ecause we can’t keep backing ambitious plays if the next layer of the pyramid doesn’t exist.

So what happens?

We fund fewer deep-tech companies.
Founders feel forced to flip to Delaware or Singapore to access aligned capital.
Domestic VCs can’t follow into those rounds because of AIF restrictions.
LPs, seeing exits dry up, write smaller cheques next time.
And the cycle repeats.

This isn’t cowardice in isolation.
It’s cowardice compounded by broken incentives.

We built the system this way. Now we need to consciously unbuild it.

The regulations aren’t broken. They’re hostile by design.

If you’ve ever tried helping a deep-tech founder register a company, import a reagent, or receive a foreign grant, you know this truth:

It’s not just that the system is broken.

It’s that it treats you like a criminal by default.

Let’s take one example.
Say you’re building a novel therapeutic. You want to test your candidate in animals. You need a test license.

In any rational ecosystem, that comes after your early in-vitro or preclinical work, once you’ve found something worth scaling.

But in India?
You often can’t get that license without already having a GMP-compliant manufacturing facility.

Let me say that again:
You have to build the factory before you’re allowed to figure out if your drug even works.

Who does that help? Not founders. Not science. Not safety.
But sure, it keeps the incumbent vaccine manufacturers happy.

This is just one example. There are dozens.
Here’s what we see again and again:

• Startup incorporation that takes weeks, not hours

• GST registrations delayed by bribery, not process

• Customs clearance for lab equipment that gets blocked because someone didn’t know which Harmonized System (HSN) code a pipette falls under

• Tariffs on critical R&D reagents jacked up to 150% overnight because someone thought scientists were smuggling ethanol

• Section 68 tax audits that allow the IT department to retroactively reclassify venture investment as income, even years later

You can’t build an R&D-led economy if founders are spending more time at the MCA portal than in their labs.
You can’t make India the hub for innovation if your customs office treats every circuit board like a Trojan horse.

The real tragedy?
Every single one of these chokepoints has been raised.
To ministries. To departments. To committees.
But they stay in place because the people with the power to fix them don’t lose sleep over the people trying to build.

So here’s what we need, urgently:

• A default-to-yes framework for scientific experimentation

• Automatic GST registration on company incorporation, at least for firms declaring R&D as core activity

• Clear import exemptions for low-volume R&D reagents and scientific equipment

• Immediate abolition of Section 68 as applied to verified foreign VC investment

• A 24-hour fast-track process for any startup operating under a registered DSIR-recognized facility, biotech park, or CSIR lab

• And honestly? A customer support helpline founders can call when their PCR machine gets classified as a microwave.

These are not moonshot asks. These are baseline hygiene.

If India wants to compete in R&D, the system has to stop assuming every founder is a smuggler, a fraud, or a liability.

Trust the builders. Then verify. But at least let them build.

The LP problem is real. And urgent.

Here’s the thing most people outside the venture ecosystem don’t always get:

A VC fund is only as bold as its LPs allow it to be.
And most LPs backing Indian VC funds are not asking for deep tech.

They’re asking for fintech, consumer brands, marketplaces, SaaS, stuff that has playbooks.
Stuff with clear comps. Stuff that fits into a deck slide with logos and exit multiples.

I’ve been in rooms where a genuinely brilliant deep-tech founder gets passed over because their business can’t show “margins comparable to SaaS.”
As if building new battery chemistry or protein design software works the same way as selling B2B subscriptions to HR teams.

And it’s not just the expectations. It’s the structure.

Indian VC funds are registered as AIFs.
Which means they can only invest 25% of their corpus into foreign-incorporated companies.

Sounds fine, until you realize that a lot of Indian deep-tech founders are forced to flip early, because it’s the only way to access capital that understands them.
And then, boom: most Indian funds can’t follow on. They’re penalized for trying to stay in the game.

And even if you’re a domestic LP who wants to back a US or Singapore-domiciled Indian company, the Liberalised Remittance Scheme (LRS) caps your investment at $250,000/year, total. Across all foreign spending.

So let’s recap.

• Deep tech in India can’t raise locally at growth stage

• Flips to foreign structure to survive

• Indian VCs can’t follow on

• Indian LPs can’t deploy much into foreign vehicles

• Foreign LPs hesitate because India looks messy

…and then we wonder why nothing scales.

This is what I mean when I say the system was designed, intentionally or otherwise, to reward imitation over innovation.

If we want deep tech to thrive, we can’t just “fund more R&D.”
We need to rewire the capital stack so that risk doesn’t get stranded at Series A.

That means:

• Educating domestic LPs that optionality > predictability

• Creating feeder structures through GIFT City or otherwise

• Easing restrictions on follow-ons and foreign capital

• Giving Indian VCs the tools to follow Indian founders, wherever they incorporate

This isn’t radical. It’s overdue.

What founders can do (and already are)

If you only look at the capital side, it’s easy to feel hopeless.

But here’s what I see every week:
Founders who are building anyway.

Founders who are doing 4am mouse studies in shared labs because they couldn’t afford to set up their own wet lab yet.
Founders who are rewriting grant applications because the last one got rejected on a technicality.
Founders who figured out how to 3D print parts for a pilot reactor because customs held their shipment for six weeks.

They are doing the work. Quietly. Stubbornly. Brilliantly.
With or without us.

And they’re not waiting for the policy to get better.
They’re not waiting for growth capital to wake up.
They’re betting on their science, their teams, and their timelines.

Our job (if we say we’re early-stage investors who believe in India’s future) is simple:
Buy them time. Buy them trust. Buy them signal.

Sometimes that’s a ₹1 crore cheque.
Sometimes it’s helping them navigate the fog of regulatory paperwork.
Sometimes it’s just telling them, “Yes, this is hard. No, you’re not crazy. Keep going.”

If you're a VC reading this, here’s an uncomfortable truth:
You may not be able to help your deep-tech founders raise their Series B right now.
But you can help them survive long enough to get there.

And if we do this well, if we give them enough time and breathing room, those same founders will become the proof point the ecosystem needs.

Founders are already doing their part.
We need to make sure they don’t have to be the ecosystem and the entrepreneur at the same time.

What early-stage capital needs to remember

In India, early-stage investing isn’t just financial. It’s infrastructural.

When we back deep-tech founders, we’re not just buying equity in a company.
We’re underwriting the very possibility that the thing they’re building can exist here.

The best early-stage funds I know understand this.
They show up for the 6-month animal trial where nothing “happens.”
They help translate pitch decks into regulatory forms.
They stay in the WhatsApp group even when timelines slip by 18 months.

Because this isn’t SaaS. This is science.
And science needs time, context, and conviction.

That doesn’t mean throwing money into the void.
It means knowing how to hold risk without compressing it into something it’s not.

A lot of Indian early-stage investors still approach deep tech with a “prove-it-fast” mindset.
But deep tech doesn’t work like that. The proof is the product.
You can’t A/B test your way to a new satellite propulsion system or a cancer immunotherapy.

So we need to remember:

• The job is not to eliminate risk. It’s to sequence it.

• You don’t need 10x revenue in year one. You need technical validation and some signal.

• Valuation isn’t a milestone. It’s a narrative tool for helping the founder survive to the next checkpoint.

If we stop trying to force deep-tech founders into consumer-tech templates, we’ll stop missing the actual magic they’re building.

And that’s where government can (and must) step in too.

What government needs to stop being: a gatekeeper.

What it needs to start being: a customer.

Here’s something we forget in all the policy noise:
The single most catalytic thing a government can do is buy from its own innovators.

Not fund. Not mentor. Not “empower.”
Just buy.

This is how the U.S. built its innovation complex.
NASA bought from SpaceX. DARPA bought from biotech startups. The Department of Defense bought chips, sensors, batteries, interfaces from companies you’ve never heard of, until they scaled.

Meanwhile, in India, startups spend more time pitching government pilots than actually doing them.

If you’re a startup building for defense, space, public health, agri-tech, good luck.

You’ll probably need:

• A contact at a PSU

• A tender with a 40-page PDF

• A “proof of concept” on your own dime

• Followed by radio silence

• Followed by maybe a procurement that gets blocked because your office didn't have permanent signage during GST inspection

It’s exhausting. It’s avoidable.

Because the truth is, government spending is the most risk-tolerant capital we already have.

We just need to deploy it like it matters.

Here’s what would help:

• A procurement quota for Indian startups in deep-tech verticals: defense, biotech, space, climate, public health

• Pre-approved vendor lists that include registered DPIIT-recognized startups, not just massive corporations

• Faster RFP cycles with transparent scoring criteria

• A government version of “try before you buy” for innovation-led products: pilot, evaluate, and then place orders

• Dedicated innovation funds inside ministries that can cut cheques without six layers of approvals

You want founders to build for India? Then India has to show up as a buyer.

Not five years later. Now.

Bridging the Growth-Stage Funding Gap: The Series A Desert Is Real

Let’s talk about the part no one wants to own:
What happens after the seed money runs out.

You’ve got a deep-tech company with early validation.
There’s a working prototype, maybe even some institutional interest.
They’ve filed IP. They’ve published real science.
They’ve survived customs, compliance, and cash flow.

And then… nothing.

Because what they need next, usually ₹30–50 crore to go from proof to product, is exactly what the Indian funding ecosystem is worst at.

Most Indian growth-stage funds are structured around “late early-stage.”
They want ₹3–5 crore in MRR. Or CAC payback in 6 months.
Or “what’s your GTM flywheel?”

But if you're building something actually hard, a reactor, a new protein, an embedded vision chip, your real GTM might be "survive three validation cycles, then negotiate a manufacturing partnership with someone in Germany."

Good luck explaining that in a Series Q pitch in India.

So founders look elsewhere.
And often? They find it.
From Singapore, from sovereigns, from specialist global funds.

Which, to be clear, is great.
But it means India becomes the country where the R&D happens, not where the value accrues.
We’re doing the homework, someone else is collecting the grade.

That’s the delta we need to close.

We don’t just need early-stage courage.
We need mid-stage conviction.

What would bridging the gap actually look like?

Some suggestions, because I’m tired of people just vibing about “need more capital”:

• Blended rounds: Government-backed funds like SIDBI, or BIRAC for biotech, can match VC cheques at Series A. If the market won’t take technical risk alone, let public capital absorb the tail.

• Follow-on mandates for sovereigns: If GIC or NIIF invests in a deep-tech company, require at least one Indian fund in the round. Not to lead, just to participate and stay in the cap table.

• Special purpose sidecars: Let early-stage VCs pool pro-rata into a deep-tech Series B+ follow-on vehicle. Even if they can't lead growth rounds, they can still support their companies with aligned, trusted capital.

• Reputational underwrite: If a founder raised a government grant, cleared regulatory hurdles, filed IP, and still can’t raise a growth round, maybe the problem isn’t the startup. Maybe it’s the market. Let’s stop punishing outliers for being early.

• Corporate-led co-investment syndicates: Indian conglomerates (Tata, Mahindra, L&T, Reliance) should step in not just as customers but as co-investors, especially in strategic tech. That’s what Samsung did for Korea. That’s what SoftBank does for Japan. We can, too.

The core idea: Don’t let the baton drop just as the sprint becomes a marathon.

Because this is the moment that makes or breaks a category.

And if we build the scaffolding right, founders won’t have to choose between flipping jurisdictions and dying on home soil.

They’ll get to scale, here. And we’ll all be better for it.

The Wall Between Academia and Industry Is Killing Us

There’s a particular kind of heartbreak that happens when a founder says:

“We actually developed this inside an academic lab… but it took us nine months to get the IP.”
“Our professor is on board, but the institute doesn’t allow joint appointments.”
“We’ve published the results, but we can’t commercialize without three levels of committee sign-off.”

This happens all the time.

And it’s maddening. Because we’re not short on good research.

We’re short on ways to turn it into the thing that changes lives and economies.

India’s academic ecosystem is buzzing with talent.
IITs, IISc, IISERs, CSIR labs- world-class people doing world-class work.

But the moment someone wants to take that work out of the lab and into the world, the friction starts: Legal ambiguity. Tech transfer delays. Bureaucratic veto power over what’s essentially a bet on ambition.

And the worst part? The system is designed to preserve credit, not create value.

So we get white papers, not products.
We get citations, not companies.

Meanwhile, other countries, China, the US, even Singapore, have industrialized their research pipelines.
Professors launch companies with their PhD students.
IP is licensed on favorable, non-exclusive terms.
Startups can run wet lab experiments inside university spaces without needing to lease an entire facility.

In India, a lot of that still feels… indulgent. Or worse, suspect.

And until we fix this, deep tech won’t scale. Because deep tech starts in the lab.

So what do we do?

Some ideas that should be obvious by now:

• Automatic IP licensing frameworks: Every government-funded academic institution should publish a standard IP transfer/royalty model for startups. No more waiting for committees to “decide.”

• Founders from academia shouldn’t have to quit to build: Offer joint appointments, sabbaticals, or advisory structures for professor-founders. Make it easy for them to stay involved.

• Create translational capital inside institutions: A small ₹5–10 crore fund per IIT/IISc focused on turning IP into pilots. Not grants. Not scholarships. Just productization.

• Open-access infra: Labs, fabrication units, prototyping facilities, all of it should be available to DPIIT-registered startups at subsidized cost. If a BTech student wants to test a sensor inside an IIT facility, let them.

• Reverse fellowships: Encourage industry professionals to embed in academic labs for 6–12 months. Build muscle memory on both sides.

And please, stop treating tech transfer offices like legal departments.
They should function like matchmakers and dealmakers, not gatekeepers.

We don’t need a revolution. We need a release valve.
One that lets talent and technology flow where it needs to go, with urgency, not apology.

If we tear down the wall between research and entrepreneurship, we won’t just see more deep-tech companies.
We’ll see more Indias: the version of this country we all secretly believe in, if we let ourselves.

From Risk Aversion to Risk Literacy

Most Indian investors, institutions, and even policymakers aren’t scared of risk.
They’re scared of not understanding it.

And that’s what deep tech is: it’s opaque, lumpy, and nonlinear.
It doesn’t come with pretty dashboards or tidy MRR.
It comes with probability trees and failure modes and timelines that look like ECG graphs.

So what do we do?

We avoid it.
We dress it up as “focus.”
We say: “Let’s wait till the science is validated.”
Or: “Let’s see how this plays out in the West first.”
Or the absolute killer: “We’ll fund it once the revenue kicks in.”

Here’s the reality: If you wait for deep tech to look safe, it’s already too late.

By the time the science is validated, someone else owns the IP.
By the time the US or China proves the model, it’s prohibitively expensive to catch up.
By the time the revenue comes, the moat’s gone.

We don’t need to become reckless.
We need to become risk-literate.

That means learning to underwrite ideas where the input is R&D and the output might be years away.
It means structuring portfolios that allow for long timelines, binary outcomes, and staggered conviction.
It means building the muscle to say: “Yes, this might fail, but if it works, it changes the game.”

And it means celebrating people who try.
Because here’s the other cultural problem: we still treat failure in deep tech like it’s shameful.

But guess what?

Every chip that never made it out of fab, every antibody that didn’t bind, every launch vehicle that didn’t lift off, those are assets, not scars. They are proof that someone tried to expand the frontier.

And as a country, we need more of that.
More swing. More belief. More room to figure it out.

We talk a lot about “India’s decade.” But the real question is: what kind of decade?

One where we optimize what already exists?
Or one where we build what doesn’t yet?

The first is safe.
The second is necessary.

And the second only happens if we build a culture that backs courage with structure, and structure with conviction.

So What Now?

We’re not going to build the future by accident.
We’re going to have to fight for it: on spreadsheets, in committee meetings, inside broken portals, and through systems that treat optimism like a liability.

But here’s the good news: the fight is already on.
You can see it in the 23-year-old running ELISA assays in a shared lab in Hyderabad.
In the civil servant quietly rewriting procurement clauses so a startup can finally bid.
In the early-stage investor backing a two-person team making new alloys for grid-scale storage.
In the founder who didn’t flip their company abroad, not because it was easier to stay, but because it mattered to try.

That’s the India I work for.
The one I bet on.

And if you’ve made it this far into this post, I’m guessing it’s the India you believe in too.

So here’s what I’ll leave you with:

If you’re a founder, please keep going. You’re not imagining it. This is hard. You’re not doing it wrong. It’s just never been easy to do something new.

If you’re an early-stage investor, write that slightly uncomfortably early cheque. Be the one who says yes first. Don’t wait for the rest of the room.

If you’re a growth-stage fund, stop waiting for deep tech to fit your models. Start building models that fit deep tech.

If you’re in government, turn the dial from permission to enablement. Say yes faster. Say no with clarity. Be the builder's ally, not their bottleneck.

And if you’re an LP, an academic, a policymaker, a curious reader, just remember: you are part of this story too.
This isn’t someone else’s problem. It’s ours.

The choice in front of us is simple:
We can be a country that services the frontier.
Or we can be the country that builds it.

Let’s choose to build. And back the people bold enough to try.

A reminder: PLEASE read Soham’s brilliant piece!

P.S: We at WTFund actively looking to fund frontier and deep tech. Please apply to us!