NextFin News - As the artificial intelligence boom pushes global power grids to their breaking point, Raleigh-based startup DG Matrix announced on February 18, 2026, that it has secured $60 million in Series A funding to accelerate the deployment of its "smarter" data center power solutions. The investment round was led by Engine Ventures and featured participation from industrial heavyweights including Mitsubishi Heavy Industries and ABB, signaling a major shift in how the technology sector views electrical infrastructure. According to TechCrunch, the capital will be used to scale manufacturing of the company’s flagship Interport platform, a solid-state transformer (SST) technology designed to replace the bulky, analog power trains that have defined data centers for decades.
The timing of the raise is critical. U.S. President Trump’s administration has prioritized domestic energy independence and the rapid expansion of AI capabilities, yet hyperscalers like Microsoft and Google are increasingly hitting a "power wall." Traditional data center power architectures rely on a fragmented chain of oil-filled transformers, rectifiers, and uninterruptible power supplies (UPS). These legacy systems are not only space-intensive but also struggle with the bidirectional power flows required by modern facilities that integrate on-site solar, battery storage, and grid-interactive services. DG Matrix claims its Interport system can achieve efficiency levels of 95% to 98%, a significant leap from the 82% to 90% efficiency typical of conventional power chains.
The core innovation lies in the transition from electromagnetic induction to high-frequency power electronics. By using silicon carbide semiconductors, DG Matrix collapses multiple functions—transformation, rectification, and inversion—into a single, software-defined device. This "power router" approach allows data centers to juggle electricity from diverse sources in real-time. For instance, a facility could prioritize on-site solar during peak generation, switch to battery reserves during grid price spikes, and maintain a seamless DC output for high-density AI racks, all without the millisecond-level interruptions that can crash a multi-million dollar training run.
This technological shift is necessitated by the sheer density of modern compute. While historical data center racks operated in the 5–15 kW range, Nvidia’s latest Blackwell and Vera Rubin architectures are pushing densities toward 100 kW per rack. At this scale, the physical footprint of traditional power gear becomes a liability. DG Matrix’s solution reportedly uses only 10% to 15% of the components found in traditional systems, drastically reducing the "white space" required for electrical rooms and allowing more floor area for revenue-generating servers. The company has already secured a supply agreement with Exowatt, a startup providing modular solar and thermal storage, to create 24/7 dispatchable power hubs for AI factories.
The competitive landscape is also heating up. On the same day as the DG Matrix announcement, Northern California-based Heron Power, led by former Tesla executive Drew Baglino, closed a $140 million Series B round for similar solid-state transformer technology. This dual-funding event underscores a pivot in venture capital toward "hard tech" infrastructure. While incumbents like Schneider Electric and Eaton have dominated the market for a century, the agility of software-defined power gives startups a unique entry point. According to industry analysts, the global data center power market is expected to grow at a CAGR of over 12% through 2030, driven almost entirely by the transition to accelerated computing.
Looking forward, the success of DG Matrix and its peers will depend on their ability to move from pilot projects to gigawatt-scale deployments. The primary hurdle remains the conservative nature of utilities and the rigorous certification processes for grid-connected hardware. However, as hyperscalers increasingly act as their own utilities—investing in modular nuclear reactors and massive behind-the-meter microgrids—the demand for intelligent power routing will likely outpace regulatory evolution. If DG Matrix can prove reliability at scale, the solid-state transformer could become the fundamental building block of the 21st-century grid, turning data centers from passive consumers into active, intelligent nodes in a decentralized energy ecosystem.
Explore more exclusive insights at nextfin.ai.
