MicrogridMedia

NAES to Operate Project Matador Gas Turbine Fleet for Fermi AI Power Campus

Fermi has selected NAES to operate and maintain the natural-gas turbine fleet at Project Matador, adding an operating partner as the Texas AI power campus moves toward first generation.

Jonas Muthoni

Published · 5 min read

Combined-cycle natural gas power plant with generation equipment and stacks in the United States. - Project Matador gas turbine fleet

Fermi Inc. selected NAES Corporation on September 28 to operate and maintain the natural-gas generation fleet at Project Matador, the company’s planned private-power and AI data-center campus in Carson County, Texas. The Project Matador gas turbine fleet is intended to supply behind-the-meter generation as Fermi develops the campus, while NAES will take responsibility for day-to-day operation and maintenance of the units as they enter service.

The agreement is an operating-readiness milestone rather than a declaration that the full power campus is online. Fermi retains ownership and control of the generation assets, while NAES will establish operating and maintenance procedures, set up work-management systems, prepare plants for service, staff the units as they come online, oversee maintenance, and handle site safety and environmental-compliance functions. Fermi has not disclosed the financial terms of the NAES contract.

NAES will operate the Project Matador gas turbine fleet

Fermi’s September 28 announcement says NAES will also bring its enterprise reliability program to the site, including operator training, asset-risk reviews, maintenance planning, control reviews and a live reliability dashboard. The companies framed the agreement around preparing the generation system before the first natural-gas units begin commercial operation.

The timing follows the arrival of Fermi’s first Siemens Energy SGT6-5000F turbines at Project Matador. Fermi reported on September 18 that the units had reached the Texas Panhandle after being transported from Germany through the Port of Houston. Their arrival established that at least part of the planned gas-generation fleet has moved beyond procurement into physical site delivery, but Fermi has not yet declared those units commercially operational.

That distinction matters because Project Matador’s headline scale remains largely a development plan. In its latest quarterly filing, Fermi describes the current campus as integrating approximately 11 GW of private-grid power generation with AI data-center infrastructure, with expansion to about 17 GW dependent on additional land transactions and permits. The Form 10-Q also says the company’s power strategy includes natural gas, grid supply, battery energy storage, solar generation and longer-term nuclear baseload.

Fermi is building an integrated private power system rather than a standalone plant

The power architecture planned for Project Matador is broader than the gas turbines NAES will operate. Fermi’s filing says battery energy storage is intended to enhance reliability and help modulate customer-facing load volatility, while solar generation is planned for energy displacement and grid power will supplement private generation. The company is also pursuing nuclear capacity for a later phase.

The U.S. Nuclear Regulatory Commission is reviewing proposed Project Matador Nuclear Units 1 through 4, a four-reactor AP1000 development intended to provide approximately 4 GW of on-site baseload generation. The NRC describes the reactors as proposed; no nuclear operating capacity should be treated as available to the campus at this stage.

This multi-resource design places Project Matador in the same broader trend as other AI infrastructure projects that are bringing generation, storage and electrical controls closer to computing loads. Microgrid Media recently reported on Energy Vault’s 1.25 GW Texas AI power agreement, which combines dispatchable generation, battery storage and grid-forming systems for hyperscale facilities rather than relying solely on conventional utility interconnection timelines.

Operations contracts are becoming part of the data-center microgrid buildout

Selecting an operator before first power is consequential because large behind-the-meter systems require utility-style operating discipline. Gas turbines, batteries, substations, protection systems and data-center loads must be coordinated continuously, and an operations organization needs procedures for maintenance, dispatch, alarms, outages, safety and environmental compliance before a campus reaches steady-state service.

A similar operating model is appearing at other large AI power systems. Microgrid Media covered Veolia’s selection to operate a 350 MW AI data-center microgrid in Ohio, where on-site generation and a 430 MWh battery system are planned to serve a computing campus. Project Matador is different in ownership, scale and resource mix, but both developments show that operations and maintenance are becoming a distinct workstream alongside equipment procurement and data-center construction.

Fermi’s first announced customer is TensorWave. The company’s quarterly filing says the initial build-to-suit lease would provide 222 MW of total facility power after the final delivery phase and includes expansion options that could increase the tenant’s total facility power to 650 MW. The filing also states that the lease’s effectiveness is subject to closing conditions, including project-level financing, so the contracted figures should not be treated as operating load today.

Gas generation is part of a wider speed-to-power strategy

Large data centers are increasingly being developed on schedules that can be shorter than the time required for new transmission and utility generation. That mismatch has driven interest in temporary generation, permanent behind-the-meter plants, batteries and microgrids. Microgrid Media’s reporting on Amazon’s North Carolina data-center power permits illustrates a different version of the same challenge, with temporary bridge generation planned before permanent utility service is available.

Project Matador takes a more expansive approach by treating private generation as part of the campus’s long-term infrastructure. That can reduce dependence on a single utility-service pathway, but it also shifts fuel procurement, emissions compliance, maintenance, reserve planning and electrical coordination into the development’s own operating model.

The September 28 NAES agreement does not resolve those project risks, nor does it establish when the first Fermi-owned turbines will begin producing electricity. It does, however, add the organization that is expected to operate the assets once they are ready.

For the Project Matador gas turbine fleet, the next verifiable milestones will be commissioning and commercial-operation declarations for individual units, followed by operating data showing how the gas turbines, batteries, grid connections and eventually other resources work together under live AI load. Until those milestones occur, Project Matador remains a large development program moving through equipment delivery, construction and operating-readiness stages rather than a fully operating 11 GW or 17 GW power campus.

Jonas Muthoni

Jonas Muthoni

Jonas Muthoni is Editor-in-Chief of Microgrid Media, where he oversees reporting and analysis on microgrids, energy storage, distributed energy resources, data center power demand, grid modernization, resilience, renewable energy, and electricity policy and markets. His work focuses on the infrastructure, technologies, and regulatory developments reshaping the power system. As Editor-in-Chief, Jonas leads Microgrid Media’s editorial strategy and standards, including story selection, source verification, technical accuracy, and the development of original reporting and analysis. His coverage draws on regulatory filings, government records, utility and company disclosures, technical documentation, and independent industry sources to explain significant developments across the evolving energy system.