“Nothing in the plant’s operating envelope is new territory; most of it is milder than systems I was running daily,” Newton said.

The founding engineers have the history and expertise to make the type of power plant that people actually want safer and more efficient than ever before. Along with CEO and Co-Founder Ben Kellie, the Applied Atomics team is deploying clean energy with no groundwater usage that is built onsite directly for the customer, with modular growth capabilities. The plants are designed specifically to meet existing regulations, done on a timeline and at a cost that is finance friendly. Our leadership team knows how to navigate complex engineering, operations, and regulations. They have already built, tested, and operated systems that meet or exceed everything a nuclear plant requires.

Chris Newton: Two Decades Building at the Extremes

Chris Newton has spent the last 20 years in charge of engineering systems within some of the most demanding hardware and infrastructure programs in aerospace, including work for Orbital, SpaceX, Virgin Orbit, Relativity, and ABL.  As opposed to many of those projects, Applied Atomics’ plants operate within well-established engineering boundaries.

“As a responsible engineer and individual contributor, I have developed, built, and operated high-pressure, high-temperature, and hazardous systems whose pressure, temperatures, and hazardous fluid conditions meet or exceed anything in our plant – pressures ranging from vacuum to tens-of-thousands of psi, and temperatures from cryogenic (-425F) to several thousand degrees Fahrenheit,” Newton said. 

The plant’s operating envelope — 2,250 psia maximum design pressure, roughly 650°F steam, and pressures as low as 3 psia in the condenser — sits well inside conditions he has routinely designed for and operated. His fluid systems experience spans helium at -425°F and 6,000 psi, subcooled liquid oxygen, RP-1, LNG, and pyrophoric TEA-TEB, which ignites on contact with air or moisture, and he has built thrusters and a supersonic jet-engine simulator reaching 2,600°F. 

It’s a similar story when it comes to scale. 

Newton joined Relativity when it had a staff of 25 and left five years later after the company had grown past 1,250 employees. He built teams that went from a single person to more than 200 team members with under 4% attrition. Those teams spanned launch engineering and operations, environmental health and safety, logistics and supply chain, infrastructure and construction, high-power electrical systems, as well as regulatory and licensing — the same subsectors Applied Atomics needs in order to stand up our power plants. 

“From a construction standpoint, the plant’s footprint is comparable to (and in most cases smaller than) the launch and test sites I have acquired, designed, and built,” Newton added. 

An Applied Atomics 100MWe power plant takes up around 17 acres. By contrast, the SpaceX’s LC-39A launch site that Newton helped with the design of and build at the Kennedy Space Center was 140 acres. 

He has managed more than $1 billion in assets and $250 million-plus budgets across launch sites, test sites, and manufacturing facilities. Those include Relativity’s Long Beach and Seattle facilities, a one-million-square-foot manufacturing site for Terran R, Relativity’s Cape Canaveral launch complex and Stennis test site, SpaceX’s launch sites at SLC-4, LC-39A, and LC-40, plus the company’s droneship and offshore landing program.

On a physical scale, Newton oversaw the cross-country transport, integration, and vertical erection of both Terran 1 and Falcon 9 rockets — effectively moving and standing up a skyscraper every couple of days. He served as the responsible party for testing the world’s largest carbon-fiber tank, a 40-foot-diameter vessel built for the original Starship, standing up a full cryogenic test capability on a barge in under a week. This is not theoretical physics in a lab. This work requires the kind of integrated systems experience and operational discipline that comes only from repeated cycles of design, build, test, and operate at significant scales.

None of this is to imply that there isn’t complexity and nuance in our nuclear energy strategy. Only to show that for engineers who have been working at extreme margins for most of their careers, the plant designs present manageable challenges, not magic. 

Paul Keutelian: The Engineer Who Turns Reactors Into Reality

Where Newton brings the operational and construction discipline, Co-Founder and CTO Paul Keutelian brings the nuclear pedigree and the systems-integration experience that make a fullstack nuclear energy deployment program possible. 

Keutelian speaks with the polite candor of his native Midwest, an approachability so disarming it buries his extensive bona fides. He holds an advanced degree in Nuclear, Plasma, and Radiological Engineering from the University of Illinois Urbana-Champaign, studying under George H. Miley, and spent three years in radiological lab operations. He was the cofounder of Radiant, where he helped design a helium-cooled, high-temperature gas reactor originally conceived for Mars and later converted for terrestrial use, writing the platform’s original nuclear and fluids physics simulation and leading its initial multiphysics modeling. His earlier fusion research put him among the early multiphysics plasma modelers working on fusion ignition and deep-space thrust and power applications — work that drew recruiting interest from the defense industrial base.

“What I’m doing is making decisions that turn designed reactors into buildable reactors and putting together the teams that can do that,” Keutelian said.

Keutelian is clear that no single person designs a reactor alone, and he doesn’t claim otherwise. His value, as he puts it, is knowing “enough to make the right calls from a more capable, competent, and qualified team” — and having built that kind of team before.

Subscribe now

After Radiant, Paul supported startups to large primes in the nuclear and aerospace industries as a consultant, working everything from engineering, design process development, team development, and business guidance.

“I am constantly interacting with all these disciplines, and most of my job is studying what everybody is doing, finding common threads and trying to give the team a cohesive direction,” Keutelian said. “And my goal is to translate these directions in their language, because a fluids person doesn’t understand an electrical person, doesn’t understand a civil person, but the choices each makes impacts these other groups, and most of my career has been diving deep with all these disciplines and learning those languages.”

In parallel, Keutelian built a career managing some of the most volatile substances and operations in aerospace, at times serving as a specialist in dangerous operations — including during early Starship tests — and assisting with Falcon anomaly investigations for other high-profile incidents such as AMOS-6, with pressures and temperatures often many times beyond anything Applied Atomics’ plants will see. He also led programs with outsized institutional reach: a design and operations guide used across hundreds of engineers, deploying a nationwide preventive maintenance program extending production maintenance to the launch sites, and an instrument calibration program bringing tens of thousands of uncalibrated devices to zero within months. 

His time at SpaceX included working with the Space Force to certify Falcon Heavy and pre-flown Falcon 9s for the National Security Space Launch program. He coordinated with the Air Force, NASA-LSP, and NRO-OSL on the multi-billion dollar contract and was part of the team spearheading the certification of Falcon 9/FH. Because of his expertise across disciplines, he breaks through historic barriers between engineers, analysts, and regulators.

Even if it’s masked by his “aw shucks” demeanor and humility, Keutelian is an actual genius. Not of the flamboyant egotist variety common at the cutting edges of chic industries, his is the rare kind of mind that can hold intricate details about completely unrelated disciplines and fields, and map in real time how distant pieces, people, and procedures need to fit together to make a complex system work. In conversation, Keutelian can switch seamlessly between fluid dynamics and novel nuclear fuels to work force dynamics and community engagement protocols. These are not siloed specialties in his conception of the project at hand: they are vertically integrated processes that need to be in dialogue with one another if things are to work like they need to.

Nowhere is this more uniquely on display than the way Keutelian speaks about the regulatory steps of selling nuclear energy to clients. Whether it’s a gesture of contrarianism for one surrounded by Silicon Valley bravado or something innately respectful in his personality, Keutelian speaks admiringly of regulators, and is more than willing to build backwards in systems design to meet what he views as their reasonable requests. 

“It’s a reframing of the whole relationship between an engineering team and regulators. What I often say is regulators embody the sum of the industry’s knowledge. They’re trying to help us, and if we learn from them properly, they make us better and faster,” Keutelian said.

This isn’t a hypothetical admiration or some marketing ploy. While at SpaceX, Keutelian grew frustrated by constantly being asked to fill out quality assurance paperwork he didn’t understand. He applied for a position on the reliability team “mostly out of spite,” got it, and built his own expertise in quality assurance and regulatory affairs. 

Keutelian’s general approach is that it’s a waste of time – sometimes a dangerous one – to try reinventing complex components others have already figured out. 

Roger Chin: Software Built for Machines That Can’t Fail

Roger Chin’s path to nuclear software runs through some of the highest-stakes codebases in aerospace and finance. He began in high-frequency options trading and was a founding member of a proprietary trading firm. There, he wrote a networking stack and trading interface that handled thousands of options updates a second at sub-millisecond latency. He went on to work in the advertising divisions at Google and Microsoft, where his team’s software shipped as part of the Xbox One, Windows 8, and Windows Phone.

From 2017 to 2020, as a senior avionics firmware engineer, Chin and one other engineer wrote the firmware for SpaceX’s Falcon 9 engine controller in six months — more than 20,000 lines of code, including a new network stack and drivers for every peripheral and IC — while passing regulatory review with the Air Force and FAA. Each controller he designed manages dozens of actuators and sensors at high frequency with fault-tolerant, redundant architecture, and the code became the technical basis for SpaceX’s subsequent avionics firmware. Later, it was repurposed for Dragon 2’s nosecone and seat controllers, as well as early Starship prototypes. He also wrote Dragon 2’s life-support gas sensor firmware and the first code to establish a laser link in space on the Starlink laser communication system. 

As Radiant’s software architect from 2020 to 2025, Chin designed the company’s simulation, modeling, and hardware-in-the-loop platforms, including SimEngine, a digital twin package that let development flow seamlessly from modeling through hardware-in-the-loop testing — the software behind Radiant’s Kaleidos digital twin and every hardware milestone through passive cooldown testing. 

Like rockets stacked atop payloads, nuclear reactors carry a degree of risk. Chin, like Applied Atomics’ other leads, designs for safety as a matter of course, not an afterthought. That means building code stacks and interfaces that people outside the company can navigate in an emergency, and structuring redundancies into programming so that there is always a safety net. 

Chin says for both rockets and nuclear reactors, “the majority of the problem isn't actually in getting it working. It's about figuring out what the vulnerabilities in the supply chain are.”

A Team That Has Already Done the Hard Part

None of these executives set out to build nuclear power plants. They wanted a new way to sell energy to customers faster, safer, and cheaper than the existing options. For many, many potential customers, that means nuclear facilities built on site in new ways that haven’t yet been tried – not because there is anything mystical or magical in nuclear fission, but because the industry has been stalled for four decades. The plant isn’t the hardest thing any of them have built — it’s simply the next one.

Share

The Applied Atomics marketing team asked CEO Ben Kellie how he has been answering these common questions. Read the Q&A with Ben below.

TLDR: The Focus should be on what determines whether nuclear projects actually move forward or stall.

Q: What problem are you solving right now?

Ben Kellie:

The world requires an immense amount of power, and the gap between demand and what can actually be delivered is widening quickly. What we’re seeing across the market is that most companies respond to that by trying to build new technology. That becomes the focus. Better reactors, new fuels, different architectures. It sounds rational, but it misses the constraint.

The issue isn’t whether we can invent something new. The issue is whether anything actually gets built. So instead of starting with invention, we start with deployment. We take high-heritage, existing technologies and focus on how to get them financed, constructed, and operated as quickly as possible. That’s the only way this scales in the timeframe the market actually needs.

Share Applied Atomics

Q: Why hasn’t someone already done this?

Ben Kellie:

Because most deep tech companies are built around the wrong starting point. Engineers found companies, which makes sense, and they build something they believe in. Then they go out and try to find a problem that fits it. You end up with very sophisticated technology looking for a reason to exist.

In infrastructure, that order breaks down. You have to start with a real constraint, something that is already creating pressure in the market, and then work backward to the technology that can solve it. If you invert that, you don’t get progress, you get delay. A lot of what looks like innovation is really just time being added to a problem that already has viable solutions.

Q: What are the real risks that could cause this to fail?

Ben Kellie:

People tend to default to technical risk, but that’s not what determines outcomes here. We’re not discovering new physics. Nuclear already works. The failure mode is almost always financial.

If you can’t raise the capital required to continue, the project stops. That can happen because capital gets pulled into less disciplined efforts, or because the market shifts, or because you can’t demonstrate a credible path to revenue early enough. But at the end of the day, it’s a financing constraint.

So, the real question isn’t whether something can work technically. It’s whether you can show, before you ask for billions of dollars, how this becomes a functioning, value-generating asset. If you can’t do that, the rest doesn’t matter.

Q: Why does this get financed when nuclear historically hasn’t?

Ben Kellie:

Historically, nuclear projects have been financed more like science experiments than infrastructure. You’re effectively assembling a collection of components and hoping they integrate into a functioning system over time. That creates long timelines and delayed returns, which capital doesn’t tolerate well.

What changes the equation is when you show up with a complete project. A defined system, a supply chain that already exists, and a clear path to revenue. You’re not asking investors to imagine how it might work. You’re showing them how it will.

Right now, the broader conditions are aligned in a way that supports this approach. There’s real demand for firm, clean power. Government financing mechanisms are available. Private capital is looking for durable, long-term assets. But alignment alone isn’t enough. You still have to demonstrate that this can be delivered in a way that avoids the cost and schedule outcomes people associate with past projects.

Share Applied Atomics

Q: Why should a customer take the risk now instead of waiting?

Ben Kellie:

There’s an assumption embedded in that question that waiting reduces risk. It doesn’t. If you’re evaluating first-of-a-kind infrastructure at all, you’ve already accepted that there’s risk involved.

What matters is how you structure it. Instead of making a binary decision, you create a pathway. You start with a letter of intent, tie it to milestones that move toward a power purchase agreement, and develop alongside other options. You maintain flexibility, but you’re also positioning yourself early in something that can actually deliver.

Over time, most of the alternatives won’t progress. That’s just the reality. The ones that do are the ones that were structured to move forward from the beginning.

Q: My community is concerned about nuclear safety. How do you address that?

Ben Kellie:

There are already 94 nuclear reactors operating in the United States today, and they provide roughly half of the country’s clean electricity. They’ve been doing that consistently, without emissions, for decades.

So, this isn’t a question of whether nuclear can operate safely. It already does, at scale. The gap is usually not in the technology itself, but in how people perceive the companies behind it and the systems they’re building.

Q: Why should communities trust you specifically?

Ben Kellie:

That skepticism is rational. If you’ve spent time in places where large industrial projects happen, especially in extraction-based economies, you’ve seen the pattern. I grew up in Alaska, and every soccer team I played on was sponsored by a company like BP or ARCO. Big companies would come in, make commitments, spend a little money locally, and then leave. People remember that.

So, when a new company shows up and says it’s going to be different, the default response should be doubt.

The only way to address that isn’t through messaging, it’s through structure. Long-term operation, sustained presence, and alignment with the communities where these assets are built. If those things aren’t real, people will see through it quickly. If they are, trust builds over time, but it comes from how the project is actually delivered, not what you say about it upfront.

Q: What’s the core shift the industry needs to make?

Ben Kellie:

Most of the industry is still focused on the question of what to build. That’s the wrong framing.

The more important question is what can actually get financed, constructed, and operated over decades. That’s a much narrower set of answers, and it tends to rely less on invention and more on disciplined execution.

If that shift doesn’t happen, we’ll keep having the same conversation ten years from now.

Leave a comment

BWXT and the mPower Reactor

BWXT is the sole manufacturer of the nuclear reactors that power the U.S. Navy’s submarine and carrier fleet, a role held by BWXT and its Babcock & Wilcox predecessor since the world’s first nuclear-powered submarine, the USS Nautilus, launched in 1954.

That fleet has run safely for seven decades. A 2025 Department of Energy report on the Naval Nuclear Propulsion Program states that the Naval Reactors team “maintains an outstanding record of over 177 million miles safely steamed on nuclear power. The Program currently operates 97 reactors and has accumulated over 7,600 reactor-years of operation. A leader in environmental protection, the Program has published annual environmental reports since the 1960s.” The reactors that fleet runs on today are manufactured by BWXT.

The capacity to manufacture reliable nuclear reactors for the U.S. Navy is considered a strategic national capability, which is governed by federal statute. The Navy does not select a vendor for these systems. Instead, it maintains a relationship with the one organization the country trusts to deliver propulsion reactors that run continuously, often submerged and in conditions that allow no margin for failure, for decades on end. BWXT has continuously earned that trust across seven decades of disciplined engineering and manufacturing.

Headquartered in Lynchburg, Virginia, BWXT employs approximately 7,000 people across more than eighty sites. The company supplies nuclear fuel, components, and technical services to the U.S. government and to the commercial nuclear power industry.

—> Partnering with BWXT has massively accelerated our wholistic plant design work, and our licensing of this reactor IP is unique in the industry.

The Technology We Have Licensed

The mPower is a 195-megawatt Generation III+ small modular reactor. Its entire nuclear steam supply system is integrated into a single pressure vessel measuring seventy-five feet tall by fifteen feet in diameter, sized deliberately to be able to ship by rail on a standard flatcar. It runs on standard commercial nuclear fuel. It does not require river or coastal siting and requires less than less than 1% of water used in conventional plants with cooling towers when producing the same amount of electricity.

The significance of the design is what it eliminates. Traditional pressurized water reactors connect their primary components with large coolant pipes, and a rupture in one of those pipes is among the most serious accident scenarios in commercial nuclear safety. In mPower, the reactor core, control rod drives, once-through steam generators, and pressurizer are located inside a single integrated pressure vessel. The most consequential failure mode in conventional commercial nuclear is, in this design, physically prevented.

The Ramp Up

The mPower program began at Babcock & Wilcox in 2008. Over the next decade, more than $500 million was invested in design development, regulatory engagement, and engineering validation. The Department of Energy began its SMR Licensing Technical Support (LTS) program in 2012 to accelerate deployment of near term SMR projects, and the first grant awarded under DOE’s LTS program was for mPower. The Tennessee Valley Authority signed a letter of intent to host the first commercial deployment at the Clinch River site in Oak Ridge, Tennessee. The U.K. National Nuclear Laboratory recommended the design for further government investigation. The American Nuclear Society and the broader nuclear engineering community received the mPower as the most pragmatic and commercially credible SMR concept of its generation.

A full-scale, 120-foot hydraulic test loop was constructed at the Center for Advanced Engineering and Research in Bedford County, Virginia. The complete data archive from the Integrated Systems Test (IST) facility tests, covering passive cooling behavior, station blackout response, decay heat removal performance, and cyber-attack resilience of the digital instrumentation and control systems, is part of BWXT’s intellectual property, which we now have exclusive access to in addition to the IST facility.

The project was shelved officially in 2017 due to market conditions, not technical issues.

That market has dramatically changed. With it has come new reactor and fuel technology development companies. However, fusion and Gen IV reactor companies don’t know how much more money they need to reach the breakthroughs required to make their technologies commercially deployable, and they don’t know the construction timeline or full costs.

—> We’re committing to this Gen III+ design because it is currently the only viable option to provide a true risk profile to lenders and investors. We offer stability, reliability, and certainty.

What the Agreement Provides

Under the agreement, Applied Atomics holds the exclusive commercial rights to deploy the mPower reactor terrestrially across the United States, Canada, and other jurisdictions with comparable nuclear liability frameworks. BWXT retains ownership of the underlying intellectual property. Applied Atomics holds the deployment rights. BWXT is the exclusive manufacturer of all new, replacement, maintenance, and overhaul components for the reactor, which allows us to leverage expertise and facilities we’d have employed no matter the reactor design we chose.

The agreement is strengthened by firm commercial and manufacturing obligations from both parties.

We are a full-stack nuclear energy company that builds, deploys, and operates power plants. Our work stands on an engineering foundation that has already absorbed the development capital, the regulatory groundwork, and the validation testing others will spend a decade producing.

—> BWXT will continue to do what BWXT does best, which is to manufacture nuclear systems to the highest standards. We’ll do what we are built to do, which is to develop sites, secure customers, finance construction, bring the plants into operation, and generate firm clean power.

What Happens Next

The first work is regulatory. The mPower program completed pre-application engagement with the Nuclear Regulatory Commission between 2009 and 2014. We re-engaged that pathway, building on the design certification work already on record. We’re targeting 2030 for our first commercial operation.

Our team is advancing site selection with a focus on industrial and technology customers whose load profiles match mPower’s modular output. The commercial market for firm, dispatchable, carbon-free power did not exist at scale when the mPower program was archived, but it exists now. Data centers, semiconductor fabrication, advanced manufacturing, and the electrification of transportation have changed the demand picture in ways that the utilities of 2015 could not have planned for, and the gas plants of today cannot adequately serve.

As we grow, we’ll share updates on site partnerships, regulatory milestones, customer commitments, and our team build-out. Speaking of team build-out —>

There’s an industry where America broke ground, led the world in discovery, innovated for three decades — and then squandered that lead over the next three decades. Accidents took lives and hurt public perception while costs soared, progress stalled, and the number of deployments fell off a cliff. The 1980’s through the early 2000’s were a complete quiet period where the industry hung on by a thread.

Sound familiar? You’d be forgiven for thinking I was talking about the nuclear industry. But, in fact, I am talking about the space launch industry. The parallels between the two are shockingly similar — except for one key difference: the space launch industry has turned around and grown massively over the last 20 years, while nuclear has not.

Space launch is the only industry where we’ve squandered our lead and recovered.
Nuclear has the opportunity to be the next.

In order to maximize the chances that the nuclear industry experiences a similar renaissance, it’s useful to understand what changed space launch (spoiler: it’s the Falcon 9) and apply those lessons. It just so happens that much of the Applied Atomics team spent the last 15 years working on just that.

Going From Rockets to Reactors

Well over a decade ago, Chris, Paul, and Ben met while working on SpaceX’s Vandenberg Launch Site where they helped take it from conception through design, build, commission, and the inaugural launch. They did everything from system design, to running crane operations in the field, to writing the procedures that operated the pad, to controlling the pad and vehicle on launch day. We were expected to wear many hats (but always steel toed boots).

When we joined, we heard all the same things we now hear about the nuclear industry:

• Engineers don’t know how to design “these things” anymore.

• There’s no technical labor pool to do the construction.

• There is no supply chain to build with.

• The regulatory environment makes the job impossible.

On the surface, the people saying these things weren’t wrong. Space had stagnated. Yet, it didn’t matter. Either way, it was our job to deliver a brand-new launch pad – the first on the west coast in decades.

Let’s look closer at each complaint to see how it might inform the Applied Atomics approach now in the nuclear industry.

“Engineers don’t know how to design these things anymore”

When we joined SpaceX, there weren’t many engineers left in the aerospace industry who possessed both the required expertise and the right mindset to succeed at a rocket startup. The latter was particularly challenging because only NASA and Boeing-Lockheed’s ULA were employing these engineers in any real number and those were definitely not hotbeds of startup thinking. Therefore, new engineers had to be produced rather than simply hired. This meant to build its workforce, SpaceX had to lean hard on young new grads.

Of course, to prevent reinventing the wheel it was important to temper that youth with the bits of crucial experience available that did meet muster. The company hired a handful of “graybeards” to corral us new grads in the right direction and save us from our worst instincts. These graybeards were central to the success of the company because they put firm boundaries on our work which prevented us chasing our tails or traversing false passes for months (or years) at a time. Constraint is healthy, and we flourished in the massive free space within those hard set boundaries. It would have been impossible to succeed with only graybeards or new grads; it was the blending of the two that drove our success.

We’re in the exact same situation today in the nuclear industry. We believe the answer now is also the same: we are bringing in critical expertise early to define the constraints of systems, build foundational processes, and prevent reliving history in the wrong ways. We’re also planning on bringing in fresh talent of all backgrounds to fill out the team, tap new energy, and break problems down to first principles. The reason this works is because no matter the industry, the physics, many of the standards, and the methodologies are the same. People have the know-how, what they’re missing is the experience that saves time and money. We can provide both.

“There’s no technical labor pool”

While the amount of massive infrastructure projects has dwindled in America, it is untrue that we do not have the labor pool required to deploy nuclear power. We dealt with the same situation in the aerospace sector – after all, few of us had experience building launch sites – but for the teams of electricians, welders, pipe fitters, and concrete workers direct past experience is not actually required.

That’s because, at their most basic, both a launch site and a nuclear power plant are nothing more than concrete, rebar, pipe, and wire. Of course, it is up to the engineers to produce designs that arrange those elements properly. And it’s up to the same team to transmute that knowledge into a set of drawings and work orders for the tech teams to follow. The technicians then need only ply their trade following the approved work packets.

Therefore, sourcing construction talent is all about tapping into the industry-wide collective expertise of our technicians. In aerospace, we pulled predominantly from oil & gas for weld teams and we will employ that same strategy here. For civil steel and concrete, we have worked with construction firms that built shipping ports, football stadiums, and shopping malls. Electricians came from large facilities such as hospitals, chemical processing plants, and other power stations. Our operators will come from other nuclear plants, US Navy Nukes, and be trained in collaboration with INL.

These are complex projects, but they aren’t inscrutable mysteries. They are elements that must be arranged according to well established industry standards, codes, and practices. America’s tradespeople know how to build.1

Crucially, Applied Atomics is building an internal Deployment team to serve as our EPC. This means we can leverage the power of a strong, lean internal team that also has aligned incentives with the rest of the company. If there’s a change, we can work it out as a team, rather than on a conference call with 50 people across five companies. This is how the SpaceX launch and landing pads we were a part of were built and we experienced first hand the ways in which this streamlines communications, compresses schedules, and lowers costs.

“There is no supply chain to build with”

It’s true that the nuclear-specific supply chain is in tough shape due to the contraction of the industry over the last three decades. However, it’s important not to artificially constrain ourselves further and to apply alternatives where we can. First and foremost, is fuel. Many companies are choosing fuels such as TRISO and HALEU that are another 5 – 7 years away from availability at scale. We use standard LEU fuel which is in use in all 94 operating reactors in the United States today, and which enjoys a robust supply chain from allied countries. Yet, even then competitors are choosing to customize that fuel to fit their application, rather than work with configurations that have already been approved. There can be benefits, however there is a penalty on cost and schedule. Striking a healthy balance is necessary.

On the components side, many components can be sourced from neighboring industries. Outside of the nuclear steam supply system (NSSS) represents the “balance of plant.” This actually makes up around 80% of the infrastructure and can enjoy high commonality with every other large-scale power plant on earth, if designed properly. By taking that approach, we ensure our plant uses the same valves, same instrumentation, same piping, same everything as every other heat-engine power plant. This is a major benefit to our program in cost savings, schedule, and supply chain resiliency.

When it comes to the reactor (and all components located on the nuclear island) the standards call for the use of “N-Stamped” components. We have some methods for addressing these issues, but we aren’t sharing them publicly yet. If you want to know more, shoot us an email and we can talk. If you want a hint, go study what SpaceX did to get NASA’s approval with the cargo straps on Dragon.

While supply chain is a challenge, and one we must keep an eye on, it is not impossible – these strategies are the same ones we used on the launch pad and in the rocket factory as well.

“The regulatory environment makes it impossible”

The regulatory environment is challenging, but our stance is that it is also helping safeguard the citizens of the US and the environment from the impacts of poor planning and careless work. It should take some work to gain permission to wield this much power among the citizenry of the United States.

In our past lives on the launch pad, we not only had multiple regulators, one of them was also often our customer. That made conditions extremely challenging, since the regulating bodies didn’t always agree with one another and sometimes didn’t agree with themselves on the customer side! The way through is with strong planning, an understanding of the rules, and a program that can tailor while working to match equivalent safety. It’s not good enough to ask for relief from a rule — we must show how we are meeting the spirit in another manner, or how that rule doesn’t apply without compromising safety. The work is on us, but it is achievable. The answer definitely is not myopic lawsuits against the regulator.2

Aligning Incentives & Minimizing Miracles

The key lesson from our collective half-century leading New Space launch and landing efforts is that cost and availability matter most. The best new technology in the world is no good if we cannot afford to deploy and use it. Applied Atomics is vertically integrating the full-stack of nuclear development and deployment because it aligns incentives and drives down cost, the same way SpaceX did with the Falcon 9. Outside engineering firms and construction management teams have different goals than the primary company. Giving those company ownership of critical systems is how you get overbudget and behind schedules like the Space Launch System in the aerospace field and Vogtle in the nuclear industry.

Space launch is the only industry where we squandered our lead and got it back. That war wasn’t won with fancy new physics, exotic fuels, or novel designs. It was won by taking existing technologies (e.g. a liquid fueled rocket with a capsule on top) and doing the hard work of making them affordable and available by vertically integrating. We believe turning around the nuclear industry will require the same approach.

That’s why our mantra is “minimize miracles,” that’s what we work on every day, and man, do we have some cool things to share soon.

Stay tuned.

First Hardware Online!

Last week, we brought our Simbox online. This box consists of four computers which work together to replicate our Powerplant control computer setup. Crucially, it also serves as the home for our Fast Sim which will provide real-time simulation of our entire coupled Powerplant. As we develop the plant, we will use a technique known as “Hardware in the Loop” or HITL to pull elements out of simulation and into real hardware (valves, pumps, regulators, fuel rods) to derisk our approach.

What does all that mean? Read on to learn more!

The Philosophy of Controls, or Controls as a Philosophy

Build the Foundation

It’s natural to wonder why this is the first thing we built. The answer is simple:

This is the work that ties all the other work together.

Our entire design philosophy is based on using modular design patterns and generating a wholistic understanding of the entire power plant in simulation as soon as possible. This unites the team across disciplines and helps them understand the impact of their work at every interface.

We applied many of the same modular design patterns we’ve used in previous hardware deployments to this plant’s control systems, and the Simbox is the test bed for those choices. It also allows us to begin employing HITL as we progress. A HITL system is not the power plant per se, but it is the connective tissue that enables the power plant to be defined, built, and operated.

By starting with the fundamental control system, we can ask ourselves “Where are the errors? Where are the unknowns?” and track and mitigate each of those via tweaking the controls responses of the power plant. We can then define the characteristic time response of systems individually and as a whole in the design by exploring these edges. Then that information is fed into our system design process iteratively, run through, and the real-world solution is put back into the sim to check its behavior and the impact on other systems.

In essence, this is a digital twin but one that we can actually manipulate, calculation within, and observe actual physics based on our real-world components. We will spend the next year validating our designs while also starting to pull them out of the digital world into the physical world via HITL.1

Where the System Isn’t

When designing a system we need to understand:

1. What does it need to do? (i.e.: what’s the mission?)

2. Can we define an approach to accomplish that? (i.e.: requirements & constraints)

3. Quantify how much margin is needed (i.e.: controls, or how wrong can I be?)

What we don’t need to know is where the system is aka the exact operating point. We only need to know that it is safely within design bounds. The load demanded by the off taker is what defines where the system is any given moment, because they are our customer. Our job is to design a power plant which is both responsive to those demands and also stays well within safety boundaries. Or put another way, the system knows where it is at all times because it knows where it isn't, which might sound like something from the Tao but is actually a hallmark of active system control.

The next step was applying these principles, so built our Simbox with four off-the-shelf Raspberry Pi’s.

The Result

We know what you’re thinking — you can’t use four Raspberry Pi’s to control a nuclear power plant.2 While we are not suggesting that our final hardware will be R-Pi’s, we have also shown over the last week that you absolutely could do it if so inclined, and we did so on purpose.

We chose the R-Pi for the Simbox because it helps us quickly, cheaply, and effectively bound our control system:

• Allows us to test the full stack of ethernet, serial connections, and input sensors.

• R-Pi’s are slow, but industrial hardware is also relatively low processor speed, so we are forced to make disciplined choices around data throughput and control rate that’s relevant to the final design.

• The R-Pi + Python’s nature makes error stack-up more obvious. We can learn how to characterize all the jitter and latency throughout the system, and design around it. The errors are larger, and if we’re able to find reliable, consistent solutions here in theory then it should be even better on the actual hardware.3

The standard approach in nuclear power has been using a expensive, long-lead, special purpose computer, or manual controls backed by analogue systems which creates as a single failure point.

We are instead using a quad-computer architecture which lets us address the Byzantine General’s Problem, exchanging information between computers reliably while complementing the plant hardware design to make the right decision without degrading performance even if one computer goes down. Much more on this soon but suffice to say it increases our reliability significantly while also dropping cost, complexity, and lead-time (i.e.: our mission in a nutshell).

Here’s how they’re running after a few days:

Four computers moving data in sync with a jitter around 30 microseconds is pretty solid for R-Pi’s and Python.

A quick note on Python. We chose it for software on this build for the same reason we chose the R-Pi’s for hardware. Ultimately, we’ll employ a lower-level language like C, an RTOS, and different hardware for guarantees of consistency and longevity but right now, to move quickly we’re isolating the need to run very fast from proving out our theory of operation. Those other options are a lot of work to implement but are known quantities. In this early phase of the design, it’s all about quantifying bounds, reducing risk, and eliminating the unknown as quickly and efficiently as possible.

Expect a similar approach to standard hardware and quick validation all based redundancy over single-point reliability with our other systems as we progress. It’s this thinking which allows us to expand the supply chain, standardize the esoteric, and bring efficiency to the fore.

Share

Subscribe now

Applied Atomics builds small modular reactor (SMR) powerplants for advanced applications like hyperscalers, e-fuels, desalination, liquified natural gas, and hydrogen production. These technologies enable the carbon-free future that Earth requires to thrive. Our team designs, builds, and operates the full stack of reactor + powerplant which are co-located on our customers’ sites for maximum uptime and flexibility. Our powerplants scale from one to ten modules which allows them to deliver 100 MWe up to 1,000 MWe per deployment.

Today, we are coming out of stealth to announce our oversubscribed $2M pre-seed round led by Aubrie Pagano and the team at Alpaca Ventures, with Unruly Capital and some very close old friends like Fred Stutzer, Zach Dunn, Jonathan Lund, SLC-4 Ventures, and MARS Partners also participating. We were fortunate to have had a highly competitive round and are honored by the belief our partners have shown in our approach.

Unlocking Nuclear Power

As you may have noticed, there is a lot of activity in the nuclear space right now. Even putting aside the massive activity in fusion, fission has been enjoying a resurgence. And that’s for a host of great reasons — nuclear power plants have the smallest footprint and lowest mining impact of any energy source. They are also latitude-agnostic, meaning they work anywhere in the universe without regard to on-the-ground conditions. Finally, they have the best capacity factor of any energy source and it’s not even close. This makes them perfect for enabling the next 1,000 years of advanced industry.

Yet we are still seeing timelines exceed a decade for new efforts. Plants take too long to deploy and often cost too much. Development of novel reactor designs and advanced fuels will undoubtedly play a part in the future, but we need to make an impact now. Already hyperscalers are being told to “bring their own power” while new utility hookups are backlogged 40% and stretching over five-year delays. Applied Atomics exists to solve these problems and meet the demand that exists now.

How We're Different

1) We’re Not Re-Inventing the Wheel

To address the power crunch, we need to move quickly. That’s why we started by surveying the existing supply chain for nuclear and power generation. We then allowed those realities to drive many of our technology choices, rather than starting with a blank piece of paper. The best new technology in the world is no good if there isn’t a supply chain to support its deployment. This country has been building powerplants for almost 150 years, so there’s no reason to reinvent the wheel on the power generation front. Make steam, spin turbines, generate power.

Similarly, we have built nuclear reactors for over 60 years. All of the operating nuclear powerplants in the United States today are light water reactors running on LEU and we have chosen to work with the Gen III+ architecture as well. We’ll be sharing more on this decision and design trade in the future but suffice to say for now that this allows us to bring well-understood architectures to the regulator, as well as opens up existing fuel supply catalogs that don’t rely on nascent supply chains or adversarial nations.

Getting early clarity on our supply chain has allowed us to complete a 30% powerplant design already this year, which identifies our major costs. From this, we modeled the first version of our power generation costs. These show us immediately competitive with the DOE’s cost goals for Advanced Nuclear for first-of-a-kind deployment, and clear line of site to competing with natural gas costs for our Nth-of-a-kind deployments.

2) We Sell Power, not Reactors

The vast majority of nuclear SMR companies develop only the reactor and leave the “balance of plant” to someone else. However, this “balance” actually represents around 80% of the infrastructure, and therefore strongly affects the final power cost. Imagine getting a new car but instead of a new vehicle arriving, a crate shows up in your driveway with an engine inside. Attached to the crate is a note that says:

“Congrats on buying your new car! Now, it’s up to you to build the ‘balance of car’ and get it licensed with the DMV and DOT.”

Even for the few SMR companies that can claim they are developing their plant, in reality these are often outsourced to large engineering design and construction firms. We believe that this adds too much risk to project timelines, adding unnecessary costs and delays as well as increased regulatory risk.

Customers need power, not components. Applied Atomics sells power through industry standard power purchase agreements, or PPAs. This little to no upfront costs for our customers, and long-term recurring revenue for our company. This also makes each project financeable through traditional means, which streamlines our future capital stack.

Our Team

Our leadership team has over 50 years of combined experience deploying launch and landing infrastructure across the New Space industry, including early on in the Falcon 9 program. We saw that F9 didn’t win the launch industry by reinventing the wheel. It utilized a traditional ground-launched liquid rocket configuration, manufactured it with simple, modern processes, and then put a capsule on top. It also used a known fuel (same as the Saturn V first stage) which relieved regulatory burden.

Crucially, the company owned the full stack of engineering, build, test, and launch which provided control over cost and schedule. Most important, Falcon 9 actually launched. A lot. And that revenue and mission data were used to unlock new efficiencies and value chains down the road.

How do we know? Much of our founding team was there, learning those lessons in real-time as engineers in design, launch, operations, and certification. Now, we see an opportunity to apply this unique perspective to the nuclear industry to address the looming power crunch.

Founders

Applied Atomics was co-founded by three seasoned entrepreneurs and leaders:

Ben Kellie (CEO) joined SpaceX in early 2012 as a Responsible Engineer helping to design, build, test, and operate the Vandenberg SLC-4E west coast launch site. He served as Lead Engineer for the inaugural flight. From there he pivoted over to Recovery and helped lead the development of the landing barges from first steel in the shipyard through multiple missions at sea. He then founded “The Launch Company” which developed launch sites for New Space rocket companies and also bootstrapped hardware to orbit. He successfully exited that company to Voyager Space in 2021.

Paul Keutelian (CTO) is an experienced systems engineer and entrepreneur with over 13 years of hands-on experience in high-stakes production hardware development. He was also an early Responsible Engineer on Vandy with ownership over four high-energy compressible, incompressible, and hazardous fluid systems at the plant scale. He then served as a lead control room operator and trainer as well as a point engineer for certification on a multi-billion-dollar launch program where he honed his ability to navigate highly regulated, precision-driven environments with three government customers. Finally, he was also previously the technical co-founder and first COO of Radiant Industries, where he led the development of the world’s first portable 1.2 MWe microreactor.

Mark Lambert (CFO) is a seasoned entrepreneur and executive who has over 17 years of experience in high-growth startups, financial management, regulatory compliance, and venture capital. He was previously the founder and CEO of a B2B SaaS platform specializing in regulatory compliance and financial accounting for highly regulated industries which processed over a billion dollars annually in financial transactions. He is also Chair of the Tech Investment Committee for NuFund Venture Group and worked closely with national venture capital funds on advanced financial modeling, and strategies for accessing and administering highly regulated government funding programs, including the U.S. Treasury's SSBIC program and the SBA's SBIC program.

Investors

We’re thrilled to have Alpaca on board, leading our pre-seed. General Partner Aubrie Pagano is joining our board and had this to say:

"We are betting that Ben and team will do for advanced industries what SpaceX did for space - be the first to bring online vertically integrated, co-located SMR power solutions. They have done first of a kind in the past, and they are a one-in-a-billion team to pull this off."

Stefano Bernardi of Unruly Capital added:

I fear what an alien species would think of us for not having abundant nuclear energy. We believe your team will remedy that!

We also had a strong coterie of former SpaceX friends who see how our unique approach can change the industry. Special shout out to Zach, Lund, David, Dan, and Christian for backing our vision.

We’re Funded & We’re on a Mission

We’re hitting the ground running. We’re already engaging with potential customers, and we’ve hired our Chief Engineer (another long-time SpaceX collaborator who we will announce soon!).

The team is deep in design to deliver our full Powerplant PDR by the end of this summer, ready to pull the trigger on long-lead hardware items with a fully defined supply chain from fuel, to reactor, to powerplant hardware. Together, these achievements will put us far ahead of the pack in understanding our realistic costs and schedule to deliver co-located power to our customers faster than any other team in history.

Want to stay in touch? Subscribe to follow along with our updates here and drop us a line on LinkedIn.

Subscribe now

Share