The ocean covers 70% of Earth’s surface. It produces 2% of our food. We have visually explored less than 0.001% of the deep seafloor—roughly the size of Rhode Island.
We have better maps of Mars than we do of our own ocean floor.
This is not a failure of curiosity. It is a failure of technology and attention. And it represents one of the largest opportunities in the world right now.
The Numbers Are Staggering
The blue economy is currently valued at somewhere between $1.5 and $3 trillion annually, depending on how you count. By 2030, it is expected to exceed $3.2 trillion. By 2050, it could quadruple. The total value of ocean assets—the natural capital sitting beneath the waves—has been estimated at $24 trillion.
These numbers are big enough to be meaningless, so let me make it concrete.
Aquaculture alone is a $313 billion market, growing 5-6% annually. It already produces 52% of global seafood. Wild fisheries hit their ceiling in the 1990s—37.7% are overfished, another 57% at maximum sustainable yield. All future growth in seafood has to come from farming. There is no alternative.
The underwater drone market is projected to grow from $5 billion to nearly $17 billion by 2034. Offshore wind is exploding. Ocean data and monitoring systems are becoming critical infrastructure for climate science, shipping, and defense.
And yet. Between 2012 and 2022, only $13 billion was invested in ocean sustainability. Total. Over a decade. That is less than what gets invested in fintech in a single quarter.
The ocean is the largest underinvested asset on the planet.
Why Nobody Builds Here
If the opportunity is so obvious, why are there not more startups? Why does ocean tech feel like a backwater compared to AI or fintech or even space?
The answer is that building for the ocean is genuinely hard.
The environment is hostile. Salt water corrodes everything. Pressure at depth is crushing—at the bottom of the Mariana Trench, it is eight tons per square inch, about a thousand times atmospheric pressure. A single failure in any component can end a mission or destroy equipment worth millions. The ocean does not forgive bad engineering.
Communication does not work. Radio waves, WiFi, GPS—none of it works underwater. You cannot just slap existing technology onto an underwater system. You have to invent new approaches for navigation, communication, and data transfer. This is not a software problem you can iterate on quickly.
The capital requirements are brutal. Hardware is expensive. Testing is expensive. Deployment is expensive. And the customer base is global and diffuse—you cannot just focus on one city or one market. You have to work with fishing operations in Norway, aquaculture farms in Chile, offshore wind installations in the North Sea, research institutions everywhere.
There are no experienced operators. Unlike climate tech, which has veterans from the cleantech 1.0 era, ocean tech is new enough that you are often pioneering without a playbook. You have to educate investors, recruit talent from adjacent fields, and figure out market dynamics that nobody has mapped.
This is why most founders avoid it. The feedback loops are slow. The iteration cycles are long. The risk of expensive failure is high. It is the opposite of the move-fast-and-break-things ethos that dominates Silicon Valley.
But this is also exactly why the opportunity is so large.
Hard Problems Create Moats
Here is what I have learned from building OctaPulse: the things that make ocean tech hard are the same things that make it defensible.
When you solve a problem that requires custom hardware, deep domain expertise, and years of iteration in a hostile environment, you build something that is very hard to replicate. You cannot just spin up a competitor in a garage. The barriers to entry are real.
Compare this to a typical SaaS product. If you build something successful, you will have competitors within months. They can copy your features, undercut your pricing, and compete for your customers with similar marketing. Your moat is your brand, your distribution, maybe your data. But the core technology is usually replicable.
In ocean tech, the technology itself is the moat. If you build a computer vision system that can accurately phenotype fish in murky underwater conditions, you have solved a problem that took years of hardware iteration, algorithm development, and domain expertise. A competitor cannot just hire some engineers and catch up. They have to go through the same painful learning process you did.
This is why deep tech companies, when they succeed, tend to succeed big. The same difficulty that slows you down also slows down everyone else.
The Aquaculture Example
I got into ocean tech through aquaculture, so let me use that as an example of what the opportunity looks like.
Fish farms today operate with less data visibility than a 1990s retail store. Most of them still sample fish manually—stressing the animals, getting inaccurate data, and spending hundreds of thousands of dollars a year on trained technicians to do work that could be automated.
Chickens have been selectively bred for decades. They grow 4x faster than they did in 1950. But only a tiny fraction of the 580+ farmed aquatic species have been through formal genetic improvement programs. Why? Because you cannot improve what you cannot measure, and measuring fish is hard when they are underwater.
This is a data problem disguised as a biology problem. The genetics are there. The potential for improvement is enormous. What is missing is the technology to capture phenotypic data at scale—non-invasively, continuously, across entire populations.
That is what we are building at OctaPulse. Computer vision systems that can measure fish in real-time, without handling them, with accuracy that exceeds human graders. This enables automated quality control, data-driven breeding programs, and operational efficiency that was previously impossible.
The technology is not magic. It is cameras, machine learning, and a lot of domain-specific engineering. But it took years to get right, and the problem required understanding both the computer vision challenges and the biological realities of fish farming.
This pattern repeats across ocean tech. The problems are not theoretically unsolvable. They just require patient, deep work that most founders are not willing to do.
Why Now
If ocean tech has always been hard, why is now the right time?
A few things have changed.
Sensors and computing have gotten dramatically cheaper. Edge AI processors like the NVIDIA Jetson make it possible to run sophisticated computer vision models on devices that can be deployed in remote, harsh environments. Ten years ago, this would have required a data center. Now it fits in a waterproof housing.
Machine learning has matured. The computer vision models we use today would have been research projects a decade ago. Deep learning has made it possible to solve perception problems that were previously intractable—like identifying individual fish in murky water, or detecting disease from subtle behavioral changes.
Climate pressure is creating urgency. The High Seas Treaty just achieved the ratifications needed to enter force in 2026. Environmental regulations are tightening. The pressure to manage ocean resources sustainably is not going away. This creates both regulatory drivers and market demand for technology that enables better monitoring and management.
Capital is finally paying attention. Funds like Propeller VC have raised dedicated ocean tech vehicles. MassChallenge runs ocean-focused accelerators. The infrastructure to support ocean startups is being built, slowly but steadily.
The window is opening. The founders who start building now will have a significant head start when the market really takes off.
What I Have Learned
Building in ocean tech has been different from anything else I have done. Slower, harder, more frustrating. But also more meaningful.
When you are working on a problem that matters—feeding the world, managing ocean resources sustainably, understanding the 70% of Earth we have barely explored—the difficulty feels worth it. The long feedback loops force you to think carefully. The hostile environment forces you to build things that actually work, not just things that demo well.
I grew up in Goa, surrounded by the ocean. I never thought that would be relevant to my career in technology. But it turns out that the thing I cared about as a kid—the sea, the fish, the way humans interact with the water—is exactly the thing I ended up building for.
Maybe that is a coincidence. Or maybe the things we care about have a way of finding us, if we pay attention.