RT-One becomes first AMPHIX AI infrastructure customer

RT-One has become the first commercial customer of the AMPHIX AI Infrastructure Platform. The deal links its planned AI infrastructure campuses in the Americas with a deployment model developed by RAVEL and Strata Expanse.

It will use AMPHIX to certify technology configurations and bring customer environments online before its campus developments are completed. The arrangement centres on an AMPHIX Centre of Excellence near Colusa, California, where RT-One plans to validate a setup it intends to repeat across multiple sites.

The announcement provides an early commercial reference for AMPHIX, which RAVEL and Strata Expanse present as a model for deploying enterprise AI infrastructure. It combines prepared land, power supply, edge networking, compute stacks and orchestration software in a single framework designed to let customers test production environments before committing to permanent builds.

At the heart of the model is the idea that companies can prove an AI deployment in a live setting rather than through a conventional pilot that differs from the final environment. According to RAVEL, these traditional pilots can take months to deploy and cost up to USD $4 million, while still failing to test the configuration ultimately used in production.

For RT-One, the immediate goal is to shorten the interval between construction and service launch. The company develops federated AI infrastructure campuses across the Americas and has already outlined a broader partnership with RAVEL around a sovereign cloud model spanning the region.

Work carried out in the California centre will certify a reference configuration for that model. Once tested, the compute, network, thermal and orchestration profile can be transferred to each RT-One site, with changes made only where local regulation requires them.

Deployment model

The Centre of Excellence is intended to let customers validate the exact environment they plan to run in production and, in some cases, begin serving their own end users from that environment while permanent facilities are still under construction. The same setting can also be used to tune models, set operating policies and train operations teams on the final production stack.

This process is especially relevant for AI infrastructure developers facing rising demand for graphics processing units, large capital commitments and pressure to meet promised service dates. In RT-One's case, it sells contracted capacity tied to firm delivery timetables and specified performance standards.

"The pace of innovation, unprecedented demand for GPUs, and complexity of AI infrastructure stacks have created an environment with more capital at risk, more operational exposure, and longer timelines before a single workload runs in production. The AMPHIX Centres of Excellence were designed to change that," said Denise Muyco, Chief Executive Officer, RAVEL.

"Customers like RT-One can validate, operationalize, and prove their infrastructure against real workloads before making major infrastructure investments. That's the model the industry needs to move towards, and RT-One is the first to run it," Muyco said.

RT-One said the centre will be used to validate and optimise its technology stack, refine policies around performance, energy use and cost through RAVEL Orchestrate AI, train operating teams and stand up customer environments ahead of full campus launch. The company expects proven configurations to become part of a standardised deployment block for later sites.

Regional rollout

The approach matters because RT-One is building across several jurisdictions. Standardising a configuration once and then reusing it across national markets may reduce the amount of repeated validation work needed at each new location.

"Our customers buy contracted capacity with a firm delivery date and defined performance standards; we pay a price if we miss either," said Fernando Palamone, Chief Executive Officer, RT-One.

"We have secured land and the energy strategy across the Americas. What the COE gives us is the ability to certify the exact configuration we intend to deploy - the same standardized block, the same stack, the same operating policies - while the campuses are under construction, rather than waiting until after they come online. Every validation hour we complete in the COE is an hour we do not spend proving the same point again in Brazil, in Paraguay, or in Colombia. That is what compresses the interval between capital commitment and revenue service: not speed on any single site, but certifying once and deploying many times," Palamone said.

The Colusa site is described as one of the largest planned AMPHIX locations. Power there will be delivered by a microgrid developed and managed by Colusa Indian Energy, adding another element to the effort to package land, energy and compute preparation into a repeatable deployment structure.

Within the AMPHIX arrangement, RAVEL is responsible for certifying the technology ecosystem, validating the deployment design and delivering operator training. Its Orchestrate AI software is used to manage workloads across the environment.

Strata Expanse supplies the physical infrastructure on which each site is built, including land, power, cooling and secure connectivity. It positions that layer as the base for customers and technology partners to deploy their compute environments.

"AMPHIX is designed to give customers a more repeatable blueprint for AI infrastructure deployment," said Greg Wood, Senior Director, Strata Expanse.

"Strata Expanse provides the physical foundation, including land, power, cooling, and secure connectivity, so customers and technology partners can focus on deploying and operating their compute environments. By validating configurations in a production-oriented setting, organizations can make future infrastructure decisions with greater speed and confidence," Wood said.