Meet The 19-Year-Old Founder Who Dropped Out of Queen's University and Raised $220,000 for His AI Company
Most people spend their first year of university figuring out where the dining hall is; Daniel Ray Edgar spent his building a company. By the end of it, Daniel was running an AI consultancy out of his dorm room that was clearing $20,000 in monthly recurring revenue, and he had taught himself every line of it. It is the kind of origin story the Bay Area loves to tell about its founders a decade after the fact. Daniel is living it now, at 20.
A first year that doubled as a startup
At 18, Daniel enrolled in Honours Computer Science at Queen's University in Canada. He arrived without a co-founder, a network, or a thesis, just a conviction that the new generation of AI tools had quietly collapsed the distance between an idea and a working product. Where an earlier cohort of student founders would have needed a team and a seed round to ship anything real, Daniel suspected that one disciplined person with the right tools could now do the work of several. He decided to test that theory on himself before he tried to sell it to anyone else.
So in his first year he taught himself to build with AI. Not in the abstract, lecture-hall sense, in the concrete sense of shipping software other people would pay for. He treated the freely available frontier models as a force multiplier and learned, by trial and error, where they were brilliant and where they quietly fell apart. That second lesson would matter enormously later.
What "building with AI" actually looked like
It is easy, in 2026, to say a young founder "used AI." It is worth being precise about what that meant, because the precision is the whole story. Daniel did not simply prompt a chatbot and paste the results. He learned to decompose a real business problem into pieces a model could handle reliably, to wire those pieces together into systems that ran without supervision, and, critically, to design around the places where the models were unreliable. He was, in effect, running a one-person engineering organization in which AI did the typing and he did the judgment.
That is a different skill from traditional programming, and it is one the universities had not yet figured out how to teach. Daniel taught it to himself the only way it can be learned: by shipping, breaking, and fixing things that someone was paying for. The freedom from legacy habits turned out to be an advantage. He never had to unlearn the slow way of doing it.
The result was Nodebase, an AI consultancy aimed at a deeply unglamorous, deeply profitable problem: businesses drowning in manual operations. Daniel did not chase the flashy end of AI. He went where the money and the pain actually were, small and mid-sized service businesses leaking revenue through broken, manual workflows.
Why real estate and mortgages
Daniel concentrated on real estate agencies and mortgage brokerages, two industries that run almost entirely on speed of follow-up and are notoriously bad at it. A lead that goes uncontacted for an hour is often a lead lost to a faster competitor. Agents and brokers are excellent at closing and terrible at the unglamorous machinery in front of the close: capturing inquiries, qualifying them, routing them, and following up relentlessly until someone books a call.
That machinery is exactly what AI automates well. Daniel built systems that captured leads the moment they arrived, qualified them automatically, and kept the conversation alive without a human having to remember to. For his clients, the effect was simple and measurable: fewer dropped leads, more booked appointments, more closed deals. For a real estate or mortgage business, that is not a productivity tweak. It is the difference between a good month and a bad one.
It worked. Nodebase grew to $20,000 in monthly recurring revenue, run entirely from a dorm-room desk. Recurring is the word that matters. Plenty of students earn money with one-off projects; very few build something that bills predictably every month while they are still taking midterms. Daniel was profitable before most of his classmates had declared a major, and he had done it without raising a dollar.
The quiet advantage of bootstrapping
In a venture ecosystem obsessed with raising, it is easy to undervalue what Daniel did first: he built a profitable company before he took a cent of outside money. Bootstrapping a service business to $20,000 a month teaches lessons that a seed round papers over. You learn to sell. You learn what customers will actually pay for, as opposed to what they say they admire. You learn to keep costs below revenue, because there is no investor cushion if you do not. And you learn to ship on a deadline that is set by payroll rather than by a board meeting.
Those are exactly the instincts that tend to be missing in founders who raise too early. Daniel acquired them before he turned 19, in the most honest classroom there is, a market that either pays you or does not. By the time he went looking for investment, he was not pitching a hope. He was pitching a track record.
Walking away from the sure thing
A $20,000-a-month business at 19 is the kind of thing most people protect with both hands. Daniel let it go. After his first year he took a year off school to build AI companies full time, betting that the larger opportunity was in building products of his own rather than deploying other people's tools for clients.
It is worth sitting with how unusual that decision is. The safe move, the move almost everyone makes, is to keep the cash flowing and treat the startup as a side project. Daniel did the opposite. He concluded that the consultancy, however profitable, was a ceiling, and that the only way to find out how high he could go was to give up the floor.
The bet resolved fast. Within three months he was selected into Antler Canada's TOR8 residency, the Toronto cohort of one of the most active early-stage investors in the world, a program built to back founders at the earliest, riskiest, most conviction-dependent stage, often before there is a product at all. Out of that residency, Daniel raised $220,000 at a $2.2M post-money valuation, at 19, for his first AI startup.
The second leap: from founder to CTO
Here is where the story stops following the usual script. Most founders who raise a first round at 19 spend the next several years defending it. Daniel walked away from his own funded company to chase a bigger thesis. He joined Finsider full time as Chief Technology Officer.
The leap only makes sense if you understand what Finsider is going after. The company is taking aim at one of the most mandatory, time-consuming, and expensive deliverables in all of mergers and acquisitions: the Quality of Earnings report. It is the document a buyer commissions before wiring hundreds of millions of dollars, the forensic check that confirms a target company's profits are real, repeatable, and not the product of accounting sleight of hand. Almost no serious deal closes without one. The work is brutally manual, it takes weeks, and senior accountants charge six figures to produce it.
Finsider's thesis is that this report can be commoditized, that the slow, artisanal, six-figure process can be rebuilt as software. As CTO, Daniel is responsible for the technical core of that bet: not bolting a chatbot onto a spreadsheet, but reconstructing how AI-native financial due diligence is performed from the ground up. He is 20 years old and, by his own description, building the future of investment banking. You can see what the company is building at finsider.ai.
The researcher behind the builder
What makes Daniel genuinely unusual in a crowd of young founders is that he writes the theory, not just the code. He is the single author of Uncertainty Propagation in Tree-Structured Language Model Reasoning, research that addresses the exact danger lurking inside any system that automates financial reasoning: error compounding. When an AI model reasons across many steps, small mistakes do not stay small, they propagate, and by the end of a long chain a slightly-off start can become a confidently-wrong conclusion. Daniel's paper formalizes how that decay works and, crucially, identifies when a tree-structured approach to reasoning defeats it. The framework was validated against four frontier models to within roughly 1%.
That is not a side hobby for a finance startup; it is the foundation of one. A diligence tool that produces a confident but subtly wrong answer is worse than useless, it is dangerous. Daniel is one of the few people building in this space who has actually studied, and published on, the precise failure mode that would sink a careless competitor.
His second paper, The Information-Maintenance Hypothesis, is more ambitious still. It is a unifying argument that aging, intelligence, and markets are the same problem in information theory, anchored on two theorems, Landauer's principle, which sets the thermodynamic cost of erasing information, and the Kelly-Cover identity, which links information to the optimal growth of capital. It is the rare piece of writing that treats a biological process, a cognitive one, and a financial one as instances of a single underlying law.
A generational pattern
Daniel is an early, unusually crisp example of a category the Bay Area is only beginning to name: the AI-native founder. This is someone who never learned the pre-AI way of building and therefore carries none of its assumptions about how big a team you need, how long an MVP takes, or how much capital it costs to find out whether an idea works. For this founder, the marginal cost of trying something has collapsed, and the binding constraint is no longer engineering capacity but taste, judgment, and the willingness to keep walking away from local maxima.
That is the thread running through Daniel's short career. A profitable consultancy was a local maximum, so he left it. A funded first startup was a local maximum, so he left that too, for a problem with a larger prize and a steeper hill. Each move traded comfort for slope.
Why the Bay Area should pay attention
Strip away the age and the headline numbers and what remains is a particular kind of operator: self-taught rather than credentialed, profitable rather than merely funded, and theoretically serious rather than just fast. Daniel built a real business before he could legally drink in the United States, walked away from it when it became a ceiling, raised money on his own terms, and then gave that up too for a harder problem.
The Bay Area has a word for people like this. It usually applies the word a few years late, after the company has scaled and the story has been sanded down into a keynote. Daniel is at the beginning of that arc, not the end of it, a 20-year-old who has already done the hard, unglamorous work of proving he can build something people pay for, and who is now pointing that ability at one of the most lucrative, most entrenched corners of finance. The smart move is to learn the name now.