An Idea Born Before the Technology Was Ready
In 2018, my wife and I won first place in a global innovation competition organized by Google, the International Olympic Committee, and Alibaba Cloud.
Our concept was called SwiMMinD.
The idea was simple to explain:
Could a machine watch a swimmer and automatically identify technical mistakes?
Today, that may sound like a natural application of artificial intelligence. In 2018, it sounded almost impossible.
The interesting part is that we were not AI developers. We were not a large technology company. We were simply people who believed that scientific progress would eventually make such a system possible.
At the time, the technology was not ready.
But the vision was.
Winning that competition did not create a product. It created an obsession.
For the next eight years, I kept coming back to the same question:
Can a machine truly understand movement in water?
Not simply record it.
Not simply measure it.
Not simply analyze a video.
Actually understand it.
That question eventually led to the creation of NEREUS, the cognitive engine behind ZLiX.
Why NEREUS?
In ancient Greek mythology, Nereus was known as the "Old Man of the Sea."
Wisdom
Nereus was known for wisdom.
Guidance
People sought his advice.
Adaptation
He had the ability to shape-shift.
That is exactly how I envision NEREUS.
Not as another artificial intelligence model.
Not as another computer vision system.
But as a cognitive engine capable of adapting and perceiving the world through different senses.
A shape-shifter.
The Problem I Discovered
When I started building ZLiX in late 2025, I finally had access to technologies that did not exist when SwiMMinD was conceived.
The results were exciting.
NEREUS could already analyze swimmers through video. In clear water, it performed surprisingly well.
But then I encountered a problem that every swimming coach already knows.
Water is rarely perfect.
The moment water becomes turbulent, visibility drops. Bubbles begin to hide arms and legs. Parts of the swimmer disappear.
When visibility decreases further, the quality of detection decreases as well. When the water becomes murky, muddy, or full of suspended particles, video alone becomes increasingly unreliable.
That made me stop and ask a very simple question:
Why are we trying to solve underwater intelligence using only vision?
Nature doesn't.
Learning from Nature
Millions of years ago, the ancestors of whales and dolphins lived on land. Over time, they adapted to life underwater.
They did not become masters of the ocean because they developed better cameras. They became masters of the ocean because they evolved better ways of perceiving their environment.
In deep or murky water, vision is often not their primary source of information.
Sound is.
The same principle can be observed in bats. A bat can pursue tiny insects in darkness, not because it sees them perfectly, but because it perceives the world differently.
The more I studied nature, the more I realized that the future of underwater intelligence may not lie in better cameras alone.
It may lie in combining multiple senses.
NEREUS the Shape-Shifter
This is where the idea of NEREUS became bigger than swimming.
A shape-shifter does not rely on one capability.
It adapts.
Sometimes NEREUS should think like a swimmer.
Sometimes NEREUS should perceive like a dolphin.
Sometimes NEREUS should navigate like a whale.
Sometimes NEREUS should detect like a bat.
Sometimes NEREUS may learn from biological adaptations found in land animals such as the cheetah.
Not because I want to imitate animals, but because evolution has already spent millions of years solving problems that engineers are only beginning to understand.
Why ignore the greatest research and development program ever created?
Nature has already paid the cost.
We simply need to learn from it.
From ZLiX to Something Much Bigger
Today, NEREUS powers ZLiX.
Its first mission is to help swimmers improve through objective analysis and structured guidance.
But that is only the beginning.
The same cognitive engine could one day help identify unhealthy fish in aquaculture. It could support marine research. It could enable autonomous systems to operate in environments where human visibility is limited.
It could help underwater vehicles understand their surroundings through multiple streams of information rather than relying on a single sensor.
That vision is also one of the reasons NEREUS is planned to become a foundation for projects such as ORCAPTOR, where we are exploring disruptive underwater intelligence and swarm technologies for maritime security and critical infrastructure protection.
Read About ORCAPTORThe Journey Has Just Begun
Many people see ZLiX and assume it is the destination.
For me, ZLiX is only the first visible manifestation of a much larger vision.
NEREUS already exists. It powers ZLiX today and is helping swimmers improve through structured analysis and guidance.
But if I am honest, I believe what exists today represents perhaps 10% of what NEREUS could ultimately become.
The current version relies primarily on video.
That is both its strength and its limitation.
As soon as water becomes turbulent, visibility decreases. Bubbles hide limbs. Reflections create noise. Murky water introduces even more uncertainty.
For years, researchers have tried to solve these challenges by building better cameras and better computer vision algorithms.
I believe the answer may lie elsewhere.
I believe the future belongs to systems that combine multiple senses, just as nature does.
From a Small Pool to the Open Ocean
Today, I conduct many of my experiments in a small swimming pool. It is where NEREUS is learning its first lessons about movement, perception, and underwater understanding.
My next goal is to secure funding and acquire the equipment needed to move beyond video alone.
I want NEREUS to learn from acoustics, sound, environmental sensing, motion, and other sources of information that nature has been using successfully for millions of years.
I also plan to transform the testing environment itself.
Swimming Pools
Controlled, clear, illuminated, and comparatively predictable.
The Ocean
Turbulent, dark, murky, acoustically complex, and constantly changing.
Eventually, I want NEREUS to learn in conditions that resemble the real underwater world: turbulence, poor visibility, suspended particles, changing light, and complex environments where traditional computer vision struggles.
My reasoning is simple.
If NEREUS can learn to perceive the world like a whale searching for squid thousands of meters below the ocean surface, then identifying a swimmer's kick despite bubbles in a swimming pool becomes a much easier problem.
The Vision
Not a better camera.
Not a better swimming application.
A cognitive system capable of understanding underwater environments by combining multiple senses into a single understanding of reality.
Today, that journey begins with swimmers.
Tomorrow, NEREUS may help aquaculture identify unhealthy fish before humans can detect problems.
It may guide autonomous underwater vehicles. It may support marine research.
It may become part of initiatives such as ORCAPTOR, where we are exploring disruptive underwater intelligence and swarm technologies for maritime security and critical infrastructure protection.
Back in 2018, when SwiMMinD won first place, the vision was ahead of the technology.
Eight years later, technology is finally starting to catch up.
And for the first time, I feel that NEREUS is standing at the shoreline, ready to take its first real steps into the ocean.
Experience the First Application of NEREUS
ZLiX is the first practical application powered by NEREUS. It provides swimmers and coaches with structured guidance for understanding swimming skills, assessing progress, and identifying what to work on next.
ZLiX is available free of charge. Try it, explore its swimming framework, and share your feedback.
Valentin Milkov
Founder of ZLiX and Maze-M
Creator of the NEREUS cognitive-engine concept
Former Bulgarian national and NCAA Division I swimmer