My father was in the hospital for weeks. They ran tests. They consulted specialists. They sent him home with no diagnosis.
A month later, they told him he had five different cancers. Five. No one had caught them—not during those weeks in the hospital, not in previous visits. The patterns existed somewhere in the system. The data that could have saved him was there. But it lived in silos, hidden behind access barriers and institutional walls.
He died because the healthcare system couldn't connect the dots.
My brother got sick with valley fever. For most people, it's minor. For him, it was devastating. But they misdiagnosed him with pneumonia. By the time he got the right diagnosis, he'd already lost over half his lung capacity. Permanent damage.
We tried to get him into Mayo Clinic. The wait times, the process, the cost—it was impossible. Specialists he saw claimed he had AIDS. He didn't. Doctors retired and passed no information on. The whole thing has been a mess for years.
Even insight from a couple of people like him at Mayo could have been the thing that gave him a normal life. Instead, the data lives in silos, hidden behind paywalls and wait times and access barriers.
I didn't know any of this would lead where it led. I just knew the system was broken, and I'd been building tools to fix broken systems.
The Training Ground
April 2025
For months before the epiphany, I was building Verve AI—what I called an "Integrated Business Intelligence Operating System." The vision was 13 specialized AI agents, each connecting to dozens of external APIs and data sources in their domain, all sharing insights across the network—not just data, but synthesized intelligence.
I called it the "Oracle Killer."
Thirteen expert agents, each understanding its domain deeply:
The key wasn't the individual agents. It was how they communicated. When one agent detected something significant, it triggered analysis across all related agents. A price increase decision would ripple through revenue projections, marketing strategy, customer service expectations, and competitive positioning—all synthesized into a unified recommendation.
I was working 12+ hours a day, building 60 database tables, 87 API endpoints, deep into vectors and embeddings and semantic representations. I was wiring my brain for distributed intelligence without knowing what it was training me to see.
The Pivot
June 16, 2025
My mother-in-law was diagnosed with colorectal cancer with liver metastases.
I stopped everything. Verve didn't matter anymore. What mattered was keeping her alive.
I started building Compass AI—a single agent to help her navigate the impossible complexity of cancer treatment. A tireless co-pilot that would research every treatment worldwide, craft insurance appeals, manage appointments, track diet considerations, prepare questions for doctors, match her to clinical trials. A compass pointing toward survival.
Not thirteen agents for business intelligence. One agent for staying alive.
The Epiphany
June 16, 2025 — Later That Evening
I was in my garage. The early morning after the diagnosis. I couldn't sleep—I was pacing between my workbench and my computer, doing what I always do when I'm deep in problem-solving: moving, thinking, iterating.
The question in my head was simple: How can I make this Compass agent more comprehensive?
I wasn't solving a specific problem. I was just thinking about what else it could do. What other features. What other ways to help.
And then I saw it.
In a split second, the Compass agent transformed. It became one of my Verve boardroom agents—but instead of business intelligence, it was survival intelligence.
Then I saw a network appear. Thousands of other cancer patients, each with their own Compass agent, all facing situations exactly like my mother-in-law's.
I saw them sharing insight in real time. Not data. Insight. Outcomes. What worked. What didn't.
I saw a breakthrough drug get introduced to the network. I watched that insight propagate to her node. Her baseline rose. And then I saw an explosion of agents—all with the same issue—propagating their survival insights, and the network baseline just kept rising.
In that split second, I knew I was seeing something that changes the world.
My face turned red. I got warm. My head felt like it expanded two sizes. It wasn't reasoning—it was observation. I was watching a complete system operate, and I knew immediately it was real because I'd already built the pieces. The Verve agents sharing insights. The semantic similarity matching. The cross-module synthesis. I'd spent months building exactly the components that would make this work.
The difference was scale. Not 13 agents. Millions. Not a boardroom. The planet.
The First Hour
I was already standing—I'd been pacing. Within moments, I shifted from what is this to how do I build this.
But within minutes, something else crystallized. I saw the whole picture—not just how it could help, but how it could be abused. I saw the regulatory implications. I saw why this needed protection.
I switched to patent mode.
Within an hour of the discovery, I sent an email to my best friends. The subject line was three words:
That was it. One equation in the body. The mathematical representation of what I'd just seen.
They probably thought I'd lost my mind. I hadn't. I'd found it.
The Vision That Won't Shut Off
It's been six months. I still see it.
Not metaphorically. Literally. The system runs in my head continuously. I can't shut it off.
What I See Now
The globe with nodes. Different colors for different domains. Healthcare networks. Agricultural networks. Automotive networks. All solving problems in real time, all baselines rising.
The propagation. A querying node reaching exactly the right agent, routed by similarity. The insight kicks back. The querying node pulses and its baseline rises. It spits back even better insight. Baselines explode across the network.
The brightness. Nodes getting brighter as they achieve better states—better health, better yield, better outcomes. And the brighter they get, the faster they help the others.
The race. Competition to curate the best patterns for any problem. Organizations racing to save the most people. It's a beautiful sight.
The initial flash was only healthcare—I saw the cancer patients sharing survival insights. Now I see all domains working simultaneously. The same vision, but overlapping networks. Healthcare. Agriculture. Autonomous vehicles. Industrial systems. All of them running in my head at once.
Tesla described something similar—running machines in his mind for weeks before building them, watching every part work. But he was visualizing single machines. I'm watching a planetary intelligence network operate across every domain, 24 hours a day, for six months.
Why I Keep Going
No one asks me what keeps me going. No one asks me anything. I've been ignored by institutions. Rejected by journals that require endorsements I can't get. Ghosted by investors. I'm financially drained from building this.
Honestly, I don't know what keeps me going except this: mathematical certainty and moral obligation.
I'm not a person who believes in the unbelievable. I'm grounded in reality, math, physics, science—things you can prove. I was already building the individual components with Verve and Compass. I know each piece works because I built it. The combination is just arithmetic.
And I have a moral obligation. My suffering now is less than the millions suffering unnecessarily from problems QIS can solve. Healthcare. Agriculture. Fraud. Industrial failures. It doesn't matter what domain—the pattern is the same.
I'm watching a system that could have saved my father. That could have helped my brother. That might still help my mother-in-law. It runs in my head constantly, and most people won't even look at the math.
But I don't have a choice. The vision is there. The math works. Someone has to build this. And I'm the one who saw it.
What I'm Asking
I'm not asking you to believe me because I had an epiphany. Epiphanies can be delusions.
I'm asking you to check the math. Every component I saw in that garage at 3 AM is proven technology. Vector embeddings. Semantic routing. DHT-based peer-to-peer networks. Local synthesis. All of it exists, all of it works at scale, all of it has been battle-tested for years.
The only question is whether combining them produces quadratic intelligence scaling. And that's not speculation—it's combinatorics. N agents that can each synthesize with matched peers create N(N-1)/2 unique opportunities. That's arithmetic.
I've run this through the most hostile expert evaluation I could construct—5 trillion to 1 starting odds, and I won. I've filed 39 provisional patents. I've written thousands of pages of documentation. I've built simulations showing R²=1.0 correlation.
The vision hasn't changed since that first flash. It's just expanded to show me all the domains simultaneously.
My father died because patterns that could have saved him were trapped in silos. My brother lost half his lungs because insight that could have helped him was hidden behind institutional walls. My mother-in-law is fighting cancer right now.
I saw the solution. I can't unsee it. And I can't stop building it.
On June 16th, 2025, I wasn't reasoning my way to QIS step by step. I saw it—complete and operating—in a single flash. Since then, I've been watching it run in my head continuously. That's not speculation. That's observation of a system that works. The math just describes what I can directly see.
Subscribe for New Articles