The Control Room of the Stars: How Fusionality is Standardizing the Fusion Power Revolution

The quest for commercial fusion energy—often described as the "holy grail" of clean power—has long been a pursuit defined by massive capital expenditures, decades-long timelines, and bespoke engineering. However, as a new wave of private startups races to bring the power of the stars to the electrical grid, a critical bottleneck has emerged: the lack of a specialized supply chain. While billions of dollars have flowed into reactor designs, the underlying infrastructure required to manage these volatile machines remains largely fragmented.

Enter Fusionality, a Lausanne-based startup that recently emerged from stealth with a $3.7 million (CHF 3 million) pre-seed funding round. Founded by Federico Felici and Jonas Buchli, the company is positioning itself as the premier architect of the "nervous system" for fusion reactors. By providing standardized, high-performance control systems and simulation environments, Fusionality aims to solve a problem that has plagued the industry for years: the inefficient, repetitive recreation of control software by every individual fusion firm.

Main Facts: Standardizing the Fusion Nervous System

Fusionality’s entry into the market marks a significant shift in the fusion ecosystem. Historically, fusion research was the domain of government-funded labs like ITER or the Max Planck Institute for Plasma Physics. Today, with over 40 private fusion companies globally, the industry is transitioning from pure scientific experimentation to industrial scaling.

The $3.7 Million Pre-Seed Raise

The recent funding round was led by Founderful and Playfair, two venture capital firms betting on the "picks and shovels" of the fusion industry. This capital injection is earmarked for the development of a suite of modular control technologies that Felici describes as "Lego blocks" for reactor management. The team, currently consisting of seven specialists, is focused on bridging the gap between theoretical plasma physics and robust industrial automation.

The Core Problem: The "80% Overlap"

According to CEO Federico Felici, the fusion industry is currently suffering from a redundancy crisis. "Across the industry, many companies make their own control systems from scratch," Felici noted. "But actually, 80% of each company’s control system is really exactly the same."

The challenge lies in the remaining 20%—the specific "language" of fusion that general-purpose industrial controllers do not speak. Fusion reactors require sub-millisecond decision-making to manage plasma, a state of matter so hot and volatile that it can damage the reactor walls if not perfectly contained by magnetic fields. Fusionality’s value proposition is to provide the standardized 80%, allowing startups to focus their resources on their unique reactor physics.

Target Market: Magnetic Confinement

Initially, Fusionality is targeting the magnetic confinement sector. This is the most mature branch of fusion technology, utilized by heavyweights such as Commonwealth Fusion Systems (CFS), Proxima Fusion, and Type One Energy. These devices use powerful superconducting magnets to trap hydrogen isotopes in a vacuum, heating them to millions of degrees until they fuse.

Chronology: From DeepMind to the Marketplace

The genesis of Fusionality is rooted in a landmark collaboration between academia and the vanguard of artificial intelligence.

The EPFL and Google DeepMind Connection

Federico Felici and Jonas Buchli met while working on one of the most significant breakthroughs in modern fusion control. Felici was a researcher at the École Polytechnique Fédérale de Lausanne (EPFL) in Switzerland, home to the Variable Configuration Tokamak (TCV). Buchli was a lead researcher at Google DeepMind.

In 2022, their collaboration resulted in a high-profile paper published in Nature, demonstrating how deep reinforcement learning (a form of AI) could be used to autonomously control the magnetic coils of a tokamak. This was a watershed moment; it proved that AI could handle the complex, non-linear dynamics of plasma more effectively than traditional manual programming.

The Realization of a Market Gap

Following the success of the DeepMind project, Felici transitioned to the AI giant full-time to continue developing simulations and machine learning interfaces. However, during their interactions with the broader fusion community, a recurring theme emerged. Founders of new fusion startups were consistently lamenting the lack of off-the-shelf components.

"We would ideally buy many of the components to make our control system, but there isn’t really anybody providing them," Felici recalled hearing from various companies. By 2024, the duo realized that the expertise they had developed at the intersection of AI and plasma physics was exactly what the commercial market required. They left their roles to found Fusionality in Lausanne, a hub for nuclear research and high-tech engineering.

Supporting Data: The Complexity of Plasma Control

To understand why Fusionality’s mission is so critical, one must look at the data regarding plasma instability.

The Microsecond Challenge

In a tokamak or stellarator, plasma must be maintained at temperatures exceeding 100 million degrees Celsius. At these temperatures, the fuel is highly unstable. A phenomenon known as a "disruption" can occur in a fraction of a second, leading to a sudden loss of confinement that can quench the fusion reaction and potentially cause mechanical stress to the reactor vessel.

• Decision Speed: Control systems must process sensor data and adjust magnetic actuators at frequencies often exceeding 10,000 times per second (10 kHz).
• Data Volume: Modern experimental reactors generate terabytes of data per day, all of which must be filtered and analyzed to maintain steady-state operation.
• Commonality: Regardless of whether a reactor is a spherical tokamak (like those built by Tokamak Energy) or a stellarator (like those by Proxima Fusion), the underlying requirements for magnetic coil power supply regulation, vacuum monitoring, and heat exhaust management are remarkably similar.

The Rise of the Fusion Supply Chain

Fusionality joins a small but elite group of "Tier 1" suppliers for the fusion age.

• Kyoto Fusioneering: Focusing on tritium breeding blankets and thermal cycle systems.
• ASG Superconductors: Providing the massive magnets required for confinement.
• Fusionality: Providing the software and hardware "brain" that coordinates these parts.

As the total private investment in fusion nears $7 billion globally, the demand for specialized suppliers is expected to grow at a CAGR of over 15% as companies move from "proof of concept" to "pilot plant" stages.

Official Responses: The Philosophy of "Lego Blocks"

The founders are clear that while AI is their background, it is not a "magic wand" for the entire industry.

On the Role of AI

Federico Felici offers a pragmatic view of artificial intelligence in the control room. "I wouldn’t be somebody who advocates for AI to take the role of controlling the entire fusion reactor," he stated. Instead, Fusionality views AI as a tool to "complement, enhance, or optimize" specific high-speed tasks, such as predicting instabilities before they occur or optimizing the shape of the plasma for maximum energy yield.

The Investor Perspective

The investors at Founderful and Playfair have signaled that their interest lies in Fusionality’s ability to act as a force multiplier for the entire sector. By lowering the barrier to entry for building a control system, Fusionality effectively reduces the "time-to-first-plasma" for new startups, accelerating the entire industry’s timeline toward commercial viability.

The "Lego" Strategy

The $3.7 million raise is being used to build a "tightly selected number of technologies." Felici’s "Lego block" analogy suggests a modular platform where a customer might buy a "Plasma Shape Control" module today and an "Instability Prediction" module tomorrow, all running on a unified, high-performance computing backbone.

Implications: Accelerating the Path to Net Zero

The emergence of companies like Fusionality has profound implications for the global energy transition and the burgeoning "Fusion Summer."

1. De-Risking the Fusion Sector

By standardizing the control architecture, Fusionality reduces the technical risk for investors backing new reactor designs. If a startup doesn’t have to "reinvent the wheel" for its control software, it can allocate more capital to its core IP—the physics of the reaction itself. This specialization is a hallmark of a maturing industry.

2. The Democratization of Fusion Research

Standardized control systems could allow smaller, more agile research teams and universities to build sophisticated experimental devices without the massive overhead of developing custom software suites. This could lead to a more diverse range of reactor designs being tested globally.

3. Economic Sovereignty and the Swiss Hub

By basing the company in Lausanne, Felici and Buchli are strengthening Switzerland’s position as a leader in high-tech nuclear engineering. With EPFL’s resources and the proximity to European fusion projects, Fusionality is well-positioned to be a central node in the European fusion ecosystem, which includes Germany’s Proxima Fusion and various UK-based initiatives.

4. Solving the Climate Equation

The ultimate implication is, of course, the timeline for carbon-free baseload power. If the "80% overlap" in control systems can be commoditized, the journey toward a commercial fusion pilot plant—currently targeted by many for the mid-2030s—becomes significantly more feasible.

Conclusion

Fusionality represents the next logical step in the evolution of nuclear energy: the transition from bespoke scientific experiments to a scalable, modular industry. By "speaking the language of fusion" and providing the essential control blocks for the reactors of tomorrow, Federico Felici and Jonas Buchli are not just building a company; they are building the infrastructure that may one day manage the most powerful energy source in the universe. As the $3.7 million pre-seed round begins to bear fruit, the industry will be watching closely to see if these "Lego blocks" can truly hold the sun in a bottle.