The anesthesiologist is asking Matt Loszak what level of unconsciousness he’d prefer while they thread a tube through his leg into his heart, and all he can think about is infinite nothingness. He’s sixteen, lying on a gurney at a hospital in Toronto, about to have tissue burned inside his heart to fix something called Wolff-Parkinson-White syndrome — too much conductive tissue. Loszak’s  heart’s electrical system occasionally short-circuits like bad wiring. The procedure involves snaking a catheter through his femoral artery, up through his body, into the chambers of his heart, and then burning away the excess tissue. The doctors call it “cardiac ablation”.

“I thought I’d be totally unconscious,” Loszak says.

The anesthesiologist explains the options. Go fully under, just partially, maybe something in between.

That night in the hospital, waiting to see if his rewired heart will keep beating properly, Loszak can’t stop thinking about what happens if it doesn’t. Not death, exactly — that’s too concrete a word. Loszak is thinking about after death. The infinite nothing that might follow.

“I used to have this thing where I’d be falling asleep,” he tells me, eighteen years later, sitting at  the prototype nuclear reactor factory he built in Austin, Texas, “and would think about what would infinite nothingness be. And I’d suddenly feel like I was in freefall, and my stomach would be sinking into my bed.”

It wasn’t anxiety, he says. “It was more of a feeling of awe.” That awe came from truly grasping the concept of “nothingness” — nothingness is not darkness, because darkness is something, but the complete absence of experience. “The sudden realization of what infinite nothingness would really be like. Never again would you ever experience anything ever again, never again would you ever talk to someone or eat food again. It just feels like it can’t be true.”

“The idea that we just happen to be alive once and never again,” he says. “That feels wrong.”

Loszak’s solution to this “problem” involves physics and probability. He’s come to believe in the multiverse — infinite universes, infinite versions of this conversation. It’s not exactly comforting, but it’s a framework that makes the singular preciousness of consciousness feel less like a cosmic accident.

“I’d much rather try and fail than not try at all,” he explains, and I can tell that he’s thinking back to that hospital room. The multiverse theory allows Loszak to believe that even if he fails in this universe, there are versions of him, in other universes, that will keep trying. And maybe, in at least one of those universes, he would succeed. The important part is not necessarily that he succeeds. It’s that he tries. This philosophy — born from staring down infinite nothingness at sixteen — would drive every major decision of his life.

The factory floor around us hums with purpose: welding stations, a water jet that can slice through six inches of metal, a towering stainless steel reactor vessel that his team assembled in just six months. On its surface, every employee has signed their name in Sharpie — a small, human gesture on a machine designed to split atoms.

For years after the heart surgery, Loszak was just building — anything, everything. He studied engineering physics at Queen’s University, he ran a house painting business that nearly tortured him. On Friday nights, while his classmates would head to the bars, Loszak would have one drink, then catch the midnight bus from Kingston to Toronto, arrive at 3 a.m., sleep a few hours at his parents’ place, then spend weekends doing door-to-door sales and estimates.  He grew the company to over 30 employees (all painters).

“To achieve those numbers, you have to do this many, paint this many homes, hire this many people, do this many estimates, and do this many cold calls,” he explains, describing the backwards math that drove him. “It’s like a funnel. You have to do 10,000 cold calls to then do a thousand of estimates, to then do 100 paint jobs.”

Why put himself through this? “Have you ever seen something like maybe a little hill or a mountain and be like, I think I can climb that? Even if you don’t like hiking, you want the dopamine of just getting to the top, of setting your sights on something and actually achieving it.”

His parents — a psychiatrist father and psychologist mother — had raised a son who approached business like a psychological experiment. “Business is really fascinating from a psychological perspective,” he says. What captivated him about his own business wasn’t the actual painting element — that was quite rote — but the people whose homes he could enter as a result. Through his venture, he found himself in the living rooms of bank executives and professors, excavating their stories. “One of my clients was an economics professor who said his greatest joy in life was failing half of his classes,” Loszak recalls, equally horrified and intrigued to this day.

The incongruity of an engineering physics student running a painting business also became his sales advantage. “People were like, oh, this person’s overqualified for painting,” he says. Trust from high-end clients came easily when Loszak was obviously slumming it.

But painting wasn’t the answer to infinite nothingness. Neither was his next venture, though it came closer to mattering.

In 2013, Loszak had an idea that would accidentally put him in competition with TikTok. Music was playing on his phone while he drove his friends on a road trip, and he thought: “It’d be cool if you just kept the music playing while you recorded a video.”

He built the app in a month and called it JamCam. “A friend thought the app was cool and he put it on Reddit that day and the post went viral and it got like 10,000 downloads in a few hours.” TechCrunch covered the app. Loszak raised $30,000 (matched by the Canadian government to make $60,000) and found a co-founder. Eventually, JamCamhit a million downloads.

“We were neck and neck with Musical.ly in the charts,” he says. Another company, Mindie — with a French team that had moved to Silicon Valley and raised millions from top VCs — had invented the exact interface TikTok would later copy wholesale.

But Loszak had a fatal flaw as a founder for this particular product. “I wanted to take videos of beautiful scenery or things,” he explains. But customers “wanted to turn the camera back on themselves and sing along selfies or do these dances and stuff. And that was happening on JamCam but I didn’t notice it, and so I didn’t leverage it and I didn’t play into it.”

The lesson was painful: “You can’t just force your vision on the world. You also have to be good at iterating with the world and figuring out what they want. And I really blew  that with JamCam.”

When he called the Mindie founder years later to commiserate about TikTok copying their interface, the founder’s response was thoughtful: “That’s the game.” Loszak’s theory as to why Musical.ly beat them both is less profound: “I think Musical.ly had some Chinese bot farms that inflated their download numbers a big way.”

By 2016, Loszak was back in his parents’ basement, nearly broke again, but now wiser. Over the past X years, he had learned how to code, learned how to fail, and learned that artistic vision without market feedback was just costly self-expression. That year, he co-founded Humi with Kevin Kleiman, who’d also “just done a failed social media app.” Humi did HR and payroll software for Canadian small businesses. Deliberately boring, but with paying customers.

“We wanted to do something with a real business model,” Loszak says.

Y Combinator luckily accepted the team (“Canadian investors are kind of like playing poker at the end of the night with a few chips. They’re just too risk averse”). And the company worked. They raised money from U.S. investors. The company grew to 150 employees. Series A, Series B.

“Y Combinator taught us that after you build your MVP and software, the company becomes the product,” he explains. “Any work you do in creating the product, you also have to do and think about creating the company because the company is the DNA that outputs more product.”

But by year four of working on HR SaaS at Humi, Loszak realized that something was wrong. “One night I was watching a movie and it was Radioactive with Marie Curie... about her discovery of radium and radioactivity. And I honestly cried during that movie because I felt like that was so much more meaningful than what I was doing.”

Loszak left Humi after the Series B, walking away from millions in unvested equity. “I still have most of my ownership, but I gave up on a part of it that was worth a few million.” And Loszak spent the next few months driving around Ontario, soul-searching, trying to figure out what he’d do next.

Somewhere along Loszak’s roadtrip, he drove past Kincardine, home to one of the world’s largest nuclear plants. The geography was loaded with meaning — he’d grown up with asthma in Toronto when “there were like 62 smog days per year.” The breathing attacks were “pretty scary. The idea of not being able to breathe as a kid was messed up.” The attacks were bad enough that “once or twice I had to go to the hospital.” When those coal plants were all turned off” he recalls,referring to Ontario’s transition to nuclear power from 2000-2010, “my asthma went away.” The realization had planted itself early: “Nuclear is pretty incredible.” But Loszak didn’t want to make another expensive mistake. “I looked at nuclear, I looked at the rest of clean tech, fusion, solar, perovskites, thermal storage, you name it. I looked at robotics, AI and biotech and everything just pushed me back to nuclear. I couldn’t stop thinking about nuclear.”

The systematic exploration was crucial. “I didn’t want to wake up and be like, “what if I’d worked on robotics?’ Now, anytime I see some flashy announcement in robotics or biotech, I’m like, ‘hey, I already went down that path. I know that nuclear is what I want to be doing.’”

Knowing what you want to do is one thing. Actually doing it (especially when “it” is as difficult as building your own nuclear reactors) is another. After returning from his roadtrip, Loszak started talking to everyone and anyone in the nuclear industry he could grab time with. He met with “100 people in nuclear” before finding his co-founder, Yasir Arafat. Arafat brought deep nuclear engineering expertise and design knowledge, while Loszak brought business acumen and startup experience.

Second, and more importantly: “We had a similar sense of humor.” The shared sensibility matters to Loszak  because “I think I spend more time with him than he does with his wife.” Together,  Loszak and Arafat began to see nuclear energy’s problem differently than anyone else Loszak had met in the industry. The drag to wider nuclear adoption was not about making reactors safer or more efficient — nuclear had a bad business model.

“Botox is more toxic than nuclear waste,” Loszak tells me, launching into one of his favorite anecdotes. “It’s the most toxic substance known to man.” He’s technically right — botulinum toxin, the neurotoxin and active ingredient behind botox, has a lower LD50 (the lethal dose required to kill 50% of a test population) than plutonium. A med spa fridge likely contains enough toxins to be “equivalent to a suitcase full of plutonium.”

The point, Loszak tells me, is that nuclear power has terrible PR. “People think a reactor equals a bomb,” he explains. If the weapons-grade stuff that goes into a nuclear bomb — highly-enriched uranium or plutonium designed to go ‘prompt-critical’ all at once — is like dynamite, then the low-enriched fuel rods that sit inside a commercial power reactor “is like a stick of incense.

“A stick of incense cannot blow up like a bomb does, because it is meant to be slowly discharged.”

Traditional nuclear’s problem was never technical — it was strategic. The industry kept building bigger reactors to chase economies of scale, but bigger also meant slower and more expensive. Bigger reactors also create more delays. When a gigawatt-scale reactor goes offline, you need expensive grid backup that can take five years to build — defeating the whole purpose of reliable baseload power. “Gigawatt scale nuclear is too slow and it has to come down for refueling and maintenance every two years.”

Loszak and Arafat’s insight: build small reactors, and then deploy them in fleets. That way, “if you’re refueling any one reactor, the rest can stay on.”

Then the AI boom hit. Suddenly, every tech giant was racing to build larger models, and massive data centers were sprouting across the country like mushrooms after a storm. Now, the biggest companies in the world needed massive amounts of reliable, clean power on demand, and their entire businesses were on the line.

“For the first time in the past 70 years since nuclear was invented, you have two things that are true, and neither of them have ever been true before,” Matt explains. “First, we have a huge amount of demand. We’re talking about demand for tens and tens of gigawatts of power in the next five years. Each gigawatt is like a whole city! When in history have you like 20, 30 cities been created in a matter of five or 10 years?”And the second factor: “These big tech companies  are willing to pay a high price for energy. They are willing to pay like 10 or 20 cents a kilowatt hour.” Traditional utilities would “laugh us out of the room” at those prices, but AI companies are “optimizing for speed, how fast can you deploy and availability.” They are far less price-sensitive than regular consumers.

To innovate in hardware, "you need a customer to help you come down the learning curve." Just as Tesla began with luxury Roadsters before making Model 3s, nuclear needs affluent early adopters to shoulder the early costs of innovation, and then drive down prices for everyone else. Traditional nuclear power takes “10 years to build a single gigawatt.” Loczak’s vision: “Can we one day build 10 gigawatts in one year?”

The historical precedent for mass industrial innovation is Henry Kaiser, a New Deal-era industrialist who famously has a hospital chain in California. But “he also built the Hoover Dam.” And “he built 1400 ships in World War II. And he took the average time to build a ship from eight months down to four days.” How did Kaiser do it? “Parallel factory manufacturing.” Instead of building one ship from start to finish, Kaiser built them in parallel —hundreds of hulls, thousands of engines, all at once. The same principle applies to reactors: instead of building one reactor at a time, build the perfect factory that produces multiple reactors at a time; Loszak and Arafat, now the co-founders of Aalo Atomics, are building hundreds of components simultaneously in a factory. The nuclear age won’t scale by going slow.

During the factory tour, Loszak tells me that  “we’re making the first full-scale prototype without nuclear fuel.” Loszak then clarifies  what “non-nuclear” means. It’s a complete reactor system — vessels, pumps, heat exchangers — minus the radioactive fuel that would make it actually generate power. This strategy of going almost all the way with manifacturing lets them test manufacturing processes, train workers, and prove the concept without regulatory complexity.

Aalo’s 40,000-square-foot facility is deliberately modest — ”our pilot line.” But in six months, Loszak and Arafat have built something remarkable. “Seven months ago we were 10 people in a WeWork,” Loszak says. “One month ago we released this. We announced, we hosted this big event where we unveiled the finished factory and the finished full scale prototype.”

Loszak’s grandfather was “an immigrant woodworker who used a lathe to make little furniture pieces.” Standing next to the industrial lathe in his new nuclear factory, Loszak notes quietly: “He would have been proud.”

The prototype reactor stands 20 feet tall, stainless steel gleaming. “We had 300 investors,” Loszak says of the unveiling. “Black curtain here... Big curtain. It’s like 20 feet tall, ten feet in diameter.”

He points to the vessel where nuclear fuel will eventually go, explaining how liquid metal will flow through the what, carrying heat to exchangers. “All the radioactive stuff will be in here,” he clarifies — not present now in the prototype, but someday. His and his teammates sharpied signatures stand out against the prototype microreactor.

I find the sight both awe-inducing and terrifying, like looking down from a mountain summit or feeling small next to a monument. How many people in the world get to impress their maker’s marks on such an act of creation? Even without the literal fuel, the team’s resolve is undeniable.

Aalo Atomics has raised a $6 million seed, a $30 million Series A, and is “getting close — fingers crossed” — to raising 100 million for the series B.” The trajectory is sharp — from two people in WeWork to a full prototype in 18 months.

“A lot of people told me nuclear was dumb when I started it,” Matt reflects, thinking back to JamCam. “But I actually feel like I have a really good gut feeling for what is going to work out well.”

The difference between now and JamCam, or Humi, is that now Loszak is not imposing his vision on the company, or just solving a real but random problem. Loszak is  responding to what the world desperately needs. AI companies want clean, reliable power at scale, and they need it now. Nuclear is “as safe as solar and wind” (by deaths per kilowatt-hour), “totally clean,” and can provide baseload power that renewables can’t match without massive batteries.

“We’re so blessed to have this customer to help us get this factory set up,” he says. Once Aalo Atomics proves out their model with data centers paying premium prices, they can pivot to serving other markets — ”utilities, desalination, industrial process heat: the sky’s the limit.”

I ask Loszak about the graveyard of failed nuclear startups. Loszak has already done the math. “What do I have to lose? I don’t want to spend my one existence at a job that I think sucks. I’d rather go try and fail.”

The heart surgery at sixteen created an awareness of meaning that Loszak has never shaken. The awareness of infinite nothingness —that stomach-dropping realization that existence could simply cease one day — became fuel for a relentless drive to create. Each company, each risk, each pivot has been another way for Loszak to assert his existence against the void. Standing in his Texas factory, surrounded by machines that will split atoms to power artificial minds, Matt Loszak has finally found a problem big enough to match his existential urgency.

“Every advanced reactor company talks about changing the world,” he says. “We’re the ones actually building.”

The universe might be infinite or singular, consciousness might be rare or present in all matter. But in this particular timeline, in this specific arrangement of atoms, Matt Loszak is building machines to make existence a little bit better.