• Enabled by SPAN’s market leading energy management and power controls technology, XFRA^TM taps into existing, underutilized power capacity in residential and small commercial spaces to deliver rapidly deployable, scalable compute power in a fraction of the time compared to new energy infrastructure buildouts.
• With the launch of XFRA, SPAN collaborates with NVIDIA, integrating their enterprise-grade technology in its offering, including one of the first-to-market uses of liquid-cooled NVIDIA RTX PRO 6000 Blackwell Server Edition GPUs.
• Collaborators include NVIDIA, Pulte Homes, and more, with initial deployments beginning later this year.

SAN FRANCISCO–(BUSINESS WIRE)–Today SPAN announced the launch of XFRA, a distributed data center solution designed to deliver gigawatts of new compute capacity amidst today’s growing power infrastructure constraints. Comprising a distributed network of compute nodes located in residential and small commercial spaces, XFRA enables both the immediate and future compute needs of hyperscalers, neoscalers and AI cloud providers. Initial launch partners include NVIDIA, the world leader in AI computing. This first-of-a-kind solution will launch with enterprise grade, liquid-cooled NVIDIA RTX PRO™ 6000 Blackwell Server Edition GPUs.

“SPAN’s unique and differentiated intellectual property in power controls enables us to improve the utilization of existing grid infrastructure,” said Arch Rao, founder and CEO of SPAN. “We have successfully deployed this capability to accelerate home electrification, unlock new home construction, and increase utility grid utilization. Now, distributed compute is the next logical extension of our technology. By building on our core strengths in power optimization and collaborating with industry leaders like NVIDIA, we are collapsing the speed-to-power gap to deliver gigawatts of cost-effective compute capacity at unprecedented speed.”

Closing the AI Speed-to-Power Gap: Accelerating Inference at Scale

AI is transforming global energy demand. In 2024, U.S. data centers consumed 183 terawatt-hours (TWh) of electricity, totaling more than 4% of America’s total electricity consumption, and experts predict it may exceed 9% by 2030. The grid infrastructure needed to support this scale can take over a decade to build, and some projects already in development have been waiting years for interconnection approval. Additionally, inference is set to account for more than half of all AI workloads by 2030, forcing hyperscalers to rethink technology designs and site selection to overcome current grid constraints. By utilizing existing grid capacity, XFRA is uniquely suited to quickly and efficiently meet industry needs for increased compute capacity.

“As the demand for AI and inference compute continues to accelerate, there is a critical need for low-latency solutions that are proximal to end users and can scale rapidly,” said Marc Spieler, Senior Managing Director of Global Energy Industry at NVIDIA. “SPAN is pioneering new ways to deploy enterprise-grade GPUs in distributed environments. The XFRA solution helps meet the specific power and latency requirements of modern inference workloads while making compute more accessible and efficient.”

Unlocking Capacity at the Grid Edge

XFRA leverages the SPAN smart electrical panel’s core built-in intelligence: integrated energy management and power controls functionality that unlock additional electrical service capacity (headroom) in the existing grid. This capacity powers high-performance compute nodes for AI inference, cloud gaming, and other AI workloads. As these demands rapidly increase, offtakers need a low-cost, low-latency solution that can scale quickly. XFRA is not intended to replace centralized data centers, but instead augment them by accelerating capacity growth at the grid edge. XFRA uniquely leverages underutilized power infrastructure in close proximity to end-users’ demand for inference compute, creating a system-wide win-win.

SPAN is working with leading homebuilders like PulteGroup to accelerate the initial rollout of XFRA on-site. “XFRA offers an innovative solution that can help to reduce build costs,” said Brian Jamison, PulteGroup VP, Strategic Sourcing & Procurement. “Building homes with SPAN Panels, XFRA, and battery backup, not only allows us to deliver homes with lower operating cost, but also allows us to use a home’s underutilized power infrastructure to benefit the grid overall.”

Multi-Stakeholder Value

XFRA delivers a win-win-win across the energy and compute ecosystem:

• Scalers: Gain immediate, flexible capacity for inference and cloud gaming without the multi-year lead times of traditional data centers.
• Homeowners: Receive a premium SPAN Panel, battery backup, and optional solar plus fixed, discounted rates for electricity and internet.
• Utilities: Can better manage peak demand and defer expensive capital expenditures by optimizing existing, underutilized grid infrastructure.

Powering the Future

By leveraging SPAN’s intelligent power orchestration, XFRA bridges the “speed-to-power” gap. This solution transforms the home into a critical node of the modern grid, meeting the urgent demand for high-performance compute while making the energy transition more affordable and resilient for everyone. With initial deployments beginning later this year, SPAN has developed a pipeline of deployment capacity to achieve gigawatt scale in 2027, enabled by XFRA’s highly distributed structure and low-friction scaling requirements. For more information, including a white paper with details on the technology architecture, visit XFRA.ai.

About SPAN

SPAN is on a mission to enable a more efficient and affordable energy future. The company began by reinventing the electrical panel and continues to transform grid-edge energy infrastructure through combined hardware-software innovation and advanced residential power control systems. Utilities, homeowners and developers all benefit from a smart, affordable and distributed electric grid. With SPAN solutions, grid operators can efficiently meet energy demand without expensive infrastructure upgrades, and those at home can manage their usage without disruption or sacrifice. Powering homes and communities with abundant, clean energy should be human-centered, technology-forward, and simply delightful. With behind and at-the-meter solutions that provide visibility and scale, SPAN helps make that possible. For more information, go to www.span.io.

For media inquiries, contact press@span.io

Source: Businesswire

• Collaboration to combine radiography, computed tomography and ultrasound capabilities to develop a full-lifecycle inspection solution for battery manufacturing, development and field service ​
• Joint inspection solution to be first in the market that can detect failure modes across the production lifecycle, providing battery manufacturers with cost efficiencies and higher yield productivity

Huerth, Germany and Emeryville, CA – March 31, 2026 – Waygate Technologies, a Baker Hughes business and global leader in non-destructive testing (NDT) solutions for industrial and energy infrastructure inspection, and Liminal Insights, a pioneer in AI-driven ultrasound inspection for batteries, today announced a strategic technology and channel collaboration to deliver the industry’s first integrated multi-modal inspection solution for battery manufacturing.

Under the terms of the agreement, the two companies will collaborate across battery manufacturing, technology development, and services inspection, including the development of next-generation probes and integrated service offerings. In addition, Waygate Technologies will be Liminal’s preferred channel and integration partner to battery gigafactory and other battery field customers, utilizing Liminal’s EchoStat inspection system across in-line inspections for high-volume battery production operations.

The goal of the joint effort is to deepen collaboration across manufacturing, technology development, and services. Both companies will explore joint product development, co-engineering of next-generation probes, and integrated service offerings for a fundamental inspection challenge that battery factories worldwide are facing: up to 90% of EV battery cells contain some form of anomaly, yet no single inspection technology can detect all failure modes across the production lifecycle.  
 

Collaboration to drive higher yield and cost efficiencies across the battery lifecycle

By bringing together the industrial computed tomography (CT) and radiography leadership of Waygate Technologies with Liminal’s real-time ultrasound AI analytics, the two companies will be able to offer a full-lifecycle inspection capability spanning cell production through to module assembly and field service.

For radiography and CT, the collaboration leverages CT systems from Waygate Technologies for quality assurance and failure analysis in laboratory environments, R&D-grade nano CT for cell development and materials characterization, and high energy CT for battery pack and module level inspection. The Waygate Technologies Krautkrämer product line adds ultrasound solutions from phased array electronics to advanced beam-forming algorithms and probes designed and purpose-built in house for industrial environments. Co-development with Liminal will combine the company’s proprietary array probe technology and signal processing electronics with Liminal’s EchoStat hardware platforms and AI analytics layer, creating a next-generation inline ultrasonic testing (UT) platform with superior sensitivity and production-line robustness.

Through the combination of multiple NDT modalities – X-ray CT as ground truth, ultrasound for inline screening, and machine learning to close the feedback loop with process data – manufacturers will gain earlier defect intervention, reduced end-of-line scrap, and data-driven process optimization.

“Even a fraction of a percent yield improvement at gigafactory scale translates to tens of millions in annual savings,” says Paul Perera, Director of Strategy, Technology & Partnerships at Waygate Technologies, a Baker Hughes business. “The goal of combining the unique strengths of both companies is to significantly increase productivity for our customers.”

“Multi-modality is the future of battery inspection. Waygate Technologies’ global installed base and deep NDT heritage make them the ideal partner to scale our EchoStat in-line inspection systems,” adds Shaurjo Biswas, CEO, Liminal Insights. “Together, we offer battery manufacturers a single-source inspection solution from R&D through volume production that no competitor can match today.”
 

Proven Results at Battery R&D Center in the UK

Waygate Technologies maintains a presence at the UK Battery Industrialisation Centre (UKBIC) in Coventry, UK – one of Europe’s leading battery process development facilities. A joint proof-of-concept program has tested over 400 battery cells through the combined UT and CT workflow, building a comprehensive battery data science dataset. The program has demonstrated that the multi-modal approach delivers over 10% cost savings versus single-modality inspection, with a further 12% improvement in defect capture rate.

The announcement of this strategic collaboration coincides with the commissioning of UKBIC’s new FIL line, providing an ideal environment to validate the integrated Waygate Technologies and Liminal inspection workflow at production-representative scale – from electrolyte filling through formation and aging – before deployment to commercial gigafactories.

Learn more and about the industrial inspection portfolio from Waygate Technologies and the partnership with Liminal:

• Waygate Technologies NDT for Batteries
• Waygate Technologies NDT for Battery and Automotive Manufacturers
• Waygate Technologies LinkedIn Channel
• Waygate Technologies YouTube Channel

The joint inspection solution is set to be the first on the market capable of detecting failure modes across the production lifecycle

Waygate Technologies’ Phoenix V|tome|x M Neo: The world‘s most flexible industrial dual-tube micro/nano CT scanner

Liminal’s EchoStat ARRAY: Ultrasonic inspection system for inline testing

Source: Baker Hughes

Normal Computing has raised $50 million in a round led by Samsung Catalyst as the startup pursues a two-pronged bet on the future of AI hardware: using AI to help semiconductor companies design chips more efficiently, while also developing a new kind of processor aimed at reducing energy use.

New investors include Galvanize, Brevan Howard Macro Venture Fund, and ArcTern Ventures, alongside existing backers Celesta Capital, Drive Capital, Eric Schmidt’s First Spark Ventures, and Micron Ventures.

CEO Faris Sbahi told Fortune the company’s software platform is already being used by more than half of the top 10 semiconductor companies by revenue, as it targets one of the industry’s biggest challenges: the rising cost and complexity of designing advanced AI chips, where even small errors can lead to expensive delays and rework.

Designing advanced AI chips has become so complex that even getting a design to “tape-out”—the point where it’s finalized for manufacturing—is increasingly prone to costly failure. Modern AI chips, which pack in tens of billions of transistors to support today’s frontier models, can cost more than $500 million to develop before a single unit ships.

Normal, founded in 2022 by former engineers and scientists from Google Brain, Google X, and Palantir, is also using its chip design software internally to build its own experimental AI hardware. It has already taped out a prototype chip using the company’s “thermodynamic” approach, which uses the inherent randomness of physical systems to compute more efficiently than traditional GPUs. It’s an early step in a longer-term effort to significantly reduce the energy demands of AI.

“The mission of the company is to go after this so-called AI energy crisis,” said Sbahi. “Data centers are expected to hit an energy wall around 2030, and most of the strategy now is to find new ways to acquire more energy—but our position is to solve the problem in terms of the hardware that we’re using.”

Seeking alternatives to existing AI hardware

Normal Computing is part of a growing group of startups exploring alternatives to conventional AI hardware, including Unconventional AI, led by former Intel AI chief Naveen Rao, which raised a $475 million seed round in January led by Andreessen Horowitz and Lightspeed Ventures. Another is Extropic, which is developing probabilistic AI chips based on a different technical approach.

Sbahi said the company chose the name “Normal Computing” to reflect its view that its approach is closer to how computation should naturally work. “We think this is the more normal way of computing,” he said, pointing to how the company’s software and hardware are designed to align with the underlying physics. “The software really matches the hardware.”

While building energy-efficient AI chips is the company’s long-term goal—initially focused on inference workloads for generative AI—the current fundraise will focus on scaling Normal’s commercial software business.

“Hopefully someday we’ll be integrated into mainstream semiconductor design manufacturing,” said Sbahi. He added that the semiconductor industry’s high costs and complexity make it difficult for new approaches to break in, which is why Normal has focused on working with existing chipmakers rather than trying to disrupt the system from the outside.

“It’s very expensive to make mistakes,” he said.

Source: Fortune