News

, a leader in embedded AI for power infrastructure, and 

, Europe’s leading AI cloud provider, today announced the deployment of the Karman AI power control platform across NexGen Cloud’s data centers.

The collaboration is designed to increase available AI compute capacity within existing grid constraints, targeting 50% additional usable capacity through intelligent GPU power orchestration and utilization of stranded energy. The initiative also strengthens infrastructure resilience by introducing real-time, AI-driven power optimisation at the rack level. Karman runs on a custom NVIDIA module and acts as the intelligence layer for next-generation power infrastructure, dynamically balancing performance, efficiency, and reliability across AI workloads.

The initial deployment is a showcase data center facility in Montreal before a broader rollout across NexGen Cloud’s European operations. For its enterprise customers, the partnership will enable NexGen Cloud’s full stack AI cloud platform, Hyperstack, to deliver more AI capacity, at a lower price, and with enhanced power-related services for its customers.

The partnership enables Hyperstack to drive superior energy performance through Karman as part of its ambitious secure private cloud expansion — including planned AI Factory deployments across North America and the Nordics, with capacity reaching hundreds of megawatts — and will extend these benefits to users of the Hyperstack on-demand compute service for AI, machine learning, and visualization applications.

“Power is the single biggest constraint for AI growth today. Billions of dollars of existing capacity are underutilized in data centers worldwide,” said Josh Brumberger, Utilidata CEO. “Karman unlocks that trapped capacity immediately while ensuring new data centers are optimized from day one. We’re excited to partner with NexGen Cloud, a neocloud that understands the urgency of maximizing today’s infrastructure and the enormous opportunity in building tomorrow’s most effectively.”

NexGen Cloud operates across Europe and North America, with infrastructure powered by 100% renewable energy across its European and Canadian data centers. NexGen Cloud, its parent company, is an NVIDIA Cloud Partner (NCP), positioning it at the forefront of European AI infrastructure. With Karman's embedded intelligence, NexGen Cloud will unlock massive capacity gains without waiting for grid upgrades or new electrical infrastructure.

"We now live in a world where 100MW+ projects are becoming commonplace. For infrastructure providers like NexGen Cloud servicing such projects, the next critical factor is energy management—specifically, maximizing the capacity already on-site. Karman gives us a significant commercial advantage by unlocking that trapped capacity; proving this level of optimization is a benefit we can then carry over directly to our end customers," said Chris Starkey, CEO of NexGen Cloud.

Utilidata’s custom module, co-developed with NVIDIA, provides the visibility, controls and safety mechanisms necessary to allow data center operators to fully utilize provisioned power capacity. By sampling at over 1 million times per second and responding with sub 20 millisecond latency, Karman sees and reacts as fast as the servers themselves, providing data-driven, software-defined safety mechanisms to deploy dramatically more GPUs within existing power infrastructure.

As data centers scale to meet AI demand, Karman complements NVIDIA's server-level optimizations with a parallel intelligence layer at the rack, row and building level. While GPUs optimize compute workloads, Karman optimizes electrical delivery, creating redundant, distributed visibility and control across both layers.

​​AI Data Center to Receive 50% Capacity Boost with AI Power Orchestration

, a pioneer in embedded AI for power infrastructure backed by NVIDIA, announced the opening of a new innovation lab that will also serve as its headquarters in Ann Arbor, Michigan, a region known as an epicenter for ingenuity and technology advancement. The new space will serve as the company’s primary hub for testing, product development, and collaboration with partners across the technology, AI, and energy sectors.

“Michigan is on the move and open for business, competing for and winning big projects in next-generation industries from cars to AI,” said 

. “Today’s announcement will create 25 new good-paying jobs and help cement Michigan as the best place for cutting-edge research and development that moves our nation forward. Let’s keep working together to show the world that Michigan is the best place to build the future.”

The move is supported by a $250,000 grant through the Michigan Business Development Program (MBDP), administered by the Michigan Economic Development Corporation (MEDC). The lab currently employs 18 staff members on site, accounting for 25% of Utilidata’s workforce, and is expected to create 25 new full-time jobs and strengthen partnerships with key local collaborators

The Ann Arbor headquarters includes dedicated distributed energy resources (DER) and data center labs that enable rapid prototyping and testing and will support the scaling of its autonomous power management platform, 

, across critical energy infrastructure. Utilidata’s team of engineers will use the expanded space to build hardware components, test new software and AI algorithms, with their main focus of testing Karman in data center servers. The data center lab currently houses NVIDIA B200 servers running AI workloads with Karman, allowing the team to measure system behavior under realistic compute conditions.

The DER lab will be configured to replicate modern electrified homes and businesses, with electric meters connected to battery storage systems, electric vehicles and rooftop solar. In this lab, the team will run demonstrations to test how Karman interacts with distributed energy devices, further refining how the platform can support grid reliability and enable utilities to better manage demands from increasing electrification.

“The focus around AI often assumes that growth requires more power. In reality, data centers already have significant unused capacity due to limitations in power infrastructure,” said 

. “Our focus is on helping operators safely and responsibly use the power they already have with high-resolution, real-time visibility and control. Our new headquarters was purpose-built to rapidly test and validate these capabilities, and to tap into Ann Arbor’s deep engineering and technology talent and accelerate our scale.”

Utilidata has maintained a presence in Michigan for several years, starting with its original 2,500-square foot innovation lab located in the Northern Brewery building, and working alongside local research institutions and technology partners to advance energy innovation. These efforts include projects with the University of Michigan’s Transportation Institute, Brooks Utility Partners, and Treetown Tech.In partnership with NVIDIA, Utilidata has built one of the smallest AI computing modules to easily embed in compact electrical devices to provide unparalleled visibility and power-flow control. The Ann Arbor headquarters will serve as a testing ground to demonstrate AI capabilities to support evolving power demands.

Utilidata’s Karman platform provides autonomous power management for AI data centers, enabling operators to better utilize the power already available to them. By embedding real-time visibility and control directly into power distribution equipment, Karman unlocks stranded capacity. The Karman platform is built on a custom NVIDIA Jetson Orin Nano, and has been embedded into smart meters to bring real-time visibility and power-flow control to the electric grid.

Utilidata Opens Ann Arbor Headquarters and Innovation Lab

, an NVIDIA-backed edge AI company, announced today the appointment of Gary Hicok as Executive Chair of the board of directors. Hicok, who recently served as an 

 to Utilidata, will transition into his new role on the board to lead alongside the senior leadership team at Utilidata. He will help scale the company’s distributed AI platform and custom NVIDIA module, Karman, for use in energy systems and critical infrastructure, including data centers.

“The rapid growth of AI factories creates a unique opportunity to unlock the value of edge computing, and there is no greater need than in data centers,” 

said Gary Hicok, Utilidata’s Executive Chair of the Board

. “As experts in power flow with a cutting-edge AI platform built on a custom, ultra-compact NVIDIA module, Utilidata is uniquely positioned to unlock up to 50% more compute capacity using existing data center infrastructure. During my time at NVIDIA, I often spearheaded expansion into new markets, and I’m excited to help Utilidata do the same as we tackle some of the most urgent energy challenges.”

As a former executive at NVIDIA for more than 25 years, Hicok has pioneered new technology and growth in edge computing. Hicok was the senior vice president of Automotive and Systems at NVIDIA, where he oversaw the NVIDIA DRIVE platforms used to develop self-driving vehicles. Before that, he served as senior vice president of NVIDIA's core logic business unit and the mobile business unit. His extensive experience will help Utilidata expand into new sectors that can benefit from edge AI, including AI factories.

“Gary Hicok has been an invaluable advisor, and we are thrilled to have him join our team as the Executive Chair of the Board,” 

 “Leveraging Gary’s experience and leadership, we’re excited to expand the impact of our Karman platform – especially in AI factories, where sub-millisecond power flow visibility, control, and guardrails paired with local AI will unlock significant power capacity.

Utilidata's Karman platform is built on a custom version of the NVIDIA Jetson Orin Nano and can be easily embedded into infrastructure devices to deliver unparalleled visibility and control of power flow. Hubbell Incorporated is the first electric hardware company to 

, creating the industry’s first AI smart meter. Like the electric grid, overprovisioning and lack of visibility and control over inefficient power use create major challenges for data centers. Karman is uniquely positioned to alleviate energy constraints and unlock significant value at the edge in both the grid and data centers.

Utilidata has built one of the smallest edge computing modules, making it easy to embed AI into critical infrastructure devices. With over a decade of experience using AI to optimize power flow on the electric distribution grid, the company’s edge AI platform – Karman, powered by their custom NVIDIA module – has already been embedded into smart meters to bring AI to the grid.

Recent Stories

Reports

Press Releases

Videos