Why Fusion Needs a Control‑System Revolution
Fusion researchers have long been building their own bespoke control rigs, reinventing the wheel for every new reactor. As Federico Felici and Jonas Buchli discovered, “80% of each company’s control system is really exactly the same,” yet no vendor speaks the language of plasma physics. The result? wasted engineering time, higher costs, and slower progress toward grid‑ready power.
Enter Fusionality, a Lausanne‑based spin‑out of Google DeepMind alumni. Their mission is simple: package the common, high‑precision hardware and software that every fusion device needs, and let innovators focus on the science instead of the circuitry.
From Idea to Funding in Record Time
Felici and Buchli turned a pain point into a business plan and secured a $3.7 million pre‑seed round from Founderful and Playfair. With Felici as CEO and Buchli as CTO, the duo is leveraging their years of hands‑on experience solving real‑world control problems for experimental tokamaks and stellarators.
Their approach dovetails with a nascent supply chain emerging around fusion power. Companies like Kyoto Fusioneering are already delivering specialized components, but the market still lacks a dedicated supplier for the “brain” of a reactor.
- Standardized control hardware reduces custom engineering effort.
- Modular software stacks can be tuned to different plasma configurations.
- Industry‑wide compatibility accelerates commercialization timelines.
What Fusionality Actually Offers
Rather than a one‑size‑fits‑all black box, Fusionality provides a toolkit that can be mixed and matched. Think of it as a LEGO set for fusion control: high‑speed digitizers, real‑time feedback loops, and AI‑ready data pipelines that speak the same protocol every team uses.
Because the underlying physics is identical across most devices—maintaining a super‑heated plasma long enough to achieve net energy gain—the same control algorithms apply. Fusionality’s hardware abstracts the low‑level electronics while their software layer offers plug‑and‑play modules for diagnostics, magnetic field shaping, and disruption mitigation.
This model promises two big wins for the industry:
- Speed: Start‑ups can go from prototype to pilot plant faster, shaving months off development cycles.
- Cost: Shared components mean lower capital expenditure and fewer custom parts to manufacture.
In a field where each megawatt of clean energy could transform the grid, those efficiencies matter.
Looking Ahead: A More Connected Fusion Ecosystem
The emergence of firms like Fusionality signals a shift from isolated labs to an interconnected supply chain. As more players adopt standardized control solutions, data sharing will improve, and collaborative AI models—something DeepMind veterans know well—could accelerate breakthroughs in plasma stability.
It won’t eliminate the hard science; it will simply give engineers the tools to iterate faster. If the industry can align on common hardware and software standards, the dream of a fusion‑powered grid may finally move from “decades away” to “within sight.”
For now, the Swiss startup is the missing piece many fusion companies have been searching for—proof that sometimes the best ideas arrive when you stop trying to reinvent the wheel and start building a better one.
Photo by Едуард Ковтонюк on Pexels
