A New Jersey fusion company rebuilding the stellarator around a simple idea: make the magnets easy to manufacture, and let software do the hard part.
For seventy years, the joke about fusion energy has been that it is always thirty years away. The physics is not the punchline - researchers have fused atoms in laboratories for decades. The punchline is the machine. Building a device that confines a plasma hotter than the sun, holds it steady, and does so cheaply enough to sell electricity has defied the field.
Thea Energy, based in an industrial park in Kearny, New Jersey, is betting that the way out is not a smarter reactor but a buildable one. The company designs stellarators - a type of magnetic-confinement fusion machine - but it has thrown out the part that made older stellarators so hard: the twisted, three-dimensional magnet coils that had to be fabricated to millimeter precision and assembled like a puzzle with no room for error.
In their place, Thea uses arrays of small, flat, high-temperature-superconductor electromagnets, licensed from the Princeton Plasma Physics Laboratory. Each one is simple. There are many of them. And instead of grinding steel into an exact shape, engineers steer the combined magnetic field with software, tuning the coils in real time to sculpt the field a plasma needs.
"We built Thea Energy to take fusion out of the lab and onto the grid," says co-founder and chief executive Brian Berzin. "Our architecture is simpler to manufacture, faster to construct, and more tolerant of real-world conditions."
It is a deliberately unglamorous pitch in a field fond of moonshots. The company machines its own parts on-site so engineers can iterate in hours rather than wait weeks on a vendor. It hires control-systems and manufacturing engineers from the New York area, not only fusion physicists. And in May 2026 it raised a $100 million Series B to prove the approach at scale.
Thea Energy's breakthroughs shift complexity from precision mechanical fabrication to software-defined controls. Gaetano Crupi · US Innovative Technology Fund
Thea's plan avoids a single hero leap. Each machine earns its keep and de-risks the next.
Arrays of small, simple, planar HTS electromagnets - licensed from PPPL - controlled as a phased array in software to shape the 3D magnetic field. Complexity lives in the control stack, not the metal.
A large-scale integrated stellarator meant to demonstrate power-plant-relevant, steady-state fusion, and to generate near-term revenue by producing radionuclides such as tritium and medical isotopes.
The planned first commercial fusion power plant, built on the same software-controlled magnets. Its "pixel-inspired" design assembles from repeatable magnet units - closer to a gas plant than a science experiment.
These magnet coils can be mass-manufactured via scalable and economical means - and we can just tune them out on the back end at the click of a button. Brian Berzin · Co-Founder & CEO
Eos is designed to sell what it makes before it sells electricity: tritium for the fusion supply chain and medical or industrial radioisotopes. That gives Thea revenue - and real operating data - years ahead of a power plant.
Helios targets utilities and grid operators needing zero-emission, around-the-clock power. Northern New Jersey sits inside PJM Interconnection - the largest U.S. electricity market - with brownfield sites near existing grid connections.
Thea plays a different game than its best-funded rivals. Commonwealth Fusion Systems (tokamaks) has raised billions; Helion is backed by Sam Altman. Thea's differentiation is not capital but capital efficiency - and a manufacturing-first design that treats "can we build a hundred of these?" as the real question. When investors toured the early lab, the common reaction was blunt: you did all this with only $20 million?
The Series B funds a second Northern New Jersey magnet-manufacturing facility, site selection and construction of Eos, and a doubling of the team - the machinery of scaling hardware, not just proving a concept.
Founded by Brian Berzin and David Gates as Princeton Stellarators, Inc.; wins two DOE INFUSE awards.
Selected as one of eight companies for the DOE Milestone-Based Fusion Development Program.
Prelude Ventures leads a capital-efficient round to advance the simplified magnet architecture.
Opens a purpose-built lab at Kearny Point and reports operating the world's first superconducting planar-coil magnet array. Previews Helios.
USIT leads a nine-figure round to scale magnet manufacturing and build Eos.
An electrical engineer who frames fusion as a manufacturing and controls challenge as much as a physics one. He argues that magnets which can be mass-produced - and tuned in software - are what turn a demo into a power plant.
A physicist with roughly 25 years at the Princeton Plasma Physics Laboratory, where he served as a department head and managing physicist and won an Edison Patent Award for the stellarator work Thea was spun out of.
Working in fusion is one of the most consequential and exciting things that one could do. Brian Berzin · Co-Founder & CEO
Thea Energy is a fusion power company developing a simplified stellarator that uses arrays of small, planar, software-controlled superconducting magnets to confine plasma - aiming to make fusion economical and mass-manufacturable.
Rather than pursuing exotic 3D magnet coils or maximum magnetic precision, Thea moves complexity into software. Its flat magnets can be mass-produced and tuned dynamically, making the machine cheaper to build and more tolerant of real-world wear.
Eos is an intermediate neutron-source stellarator meant to demonstrate power-plant-relevant fusion and earn revenue from tritium and isotopes (targeted around 2030). Helios is the planned first commercial power plant, targeting grid power around 2035.
It was founded in 2022 by CEO Brian Berzin and CTO David Gates as a spinout of Princeton Plasma Physics Laboratory, and is headquartered in Kearny, New Jersey.
At least $120 million - a $20 million Series A in 2024 and a $100 million Series B in 2026 led by the US Innovative Technology Fund - plus U.S. Department of Energy public-private partnership awards.
Profile compiled from public sources including Thea Energy, Wikipedia, POWER Magazine, Heatmap News, TechCrunch, NJBIZ and ROI-NJ. Dates and figures are approximate where noted and reflect information available as of mid-2026.