CoreWeave Inc. wants to step in before the infrastructure stack buckles under the weight of agentic artificial intelligence.

The company recently announced the industry’s first validation of Nvidia Vera Rubin NVL72 on CoreWeave Cloud. This suggests that AI infrastructure is entering a new stage, according to Rob Strechay, principal analyst for theCUBE Research, who sees CoreWeave as one of multiple components supporting the evolving agentic enterprise.

“The significance of Dell shipping Nvidia Vera Rubin-based systems to CoreWeave isn’t simply about being first,” Strechay said. “It signals the next phase of AI infrastructure, in which cloud providers, platform operators and infrastructure vendors are co-engineering entire AI factories focused on inference, data movement and operational efficiency, including the energy sustainability of deployments.”

The “Scaling the Agentic Era” event, airing June 30 at 9:30 a.m. PT/12:30 p.m. ET, will feature panels with CoreWeave, Nvidia Corp. and Dell Technologies Inc. leaders, as part of an exclusive broadcast on theCUBE, SiliconANGLE Media’s livestreaming studio. The discussion focuses on building accelerated computing infrastructure for AI and the needs of the agentic enterprise. (* Disclosure below.)

Building the agentic AI infrastructure

Neoclouds are reshaping the AI infrastructure model by providing an alternative to hyperscalers, increasing competition across the infrastructure market. CoreWeave is the best-known of the neoclouds, which focus primarily on AI workloads, and Nvidia is one of its key investors. In January, Nvidia bought an additional $2 billion in CoreWeave stock, and the recent Vera Rubin validation represents another vote of confidence from the GPU behemoth.

“CoreWeave’s successful bring-up of Nvidia Vera Rubin NVL72 reinforces its position as a leading AI infrastructure provider,” said Paul Nashawaty, principal analyst for theCUBE Research. “Early access to next-generation inference hardware, combined with differentiated rack-scale engineering, could strengthen the company’s competitive advantage as agentic AI workloads drive demand for higher-performance, lower-cost inference.”

Dell will also join an exclusive event panel with CoreWeave and Nvidia. These high-level AI collaborations reflect the complexity of AI deployment, with liquid cooling, rack control and networking being just a few of the components required to support large-scale inference. Agentic AI puts even more pressure on infrastructure.

“As organizations move from being token consumers to token producers, competitive advantage will increasingly come from how efficiently data, storage, networking and GPUs work together, not from any single component alone,” Strechay said.

CoreWeave looks ahead

CoreWeave has developed purpose-built infrastructure for Vera Rubin, including Valvey, a programmable per-rack valve assembly that turns cooling from a passive mechanical system into a software-defined control surface, and Racky, a unified rack control appliance with a standardized management surface.

The goal is for Vera Rubin to perform not just in a lab but at production scale, using CoreWeave’s full-stack orchestration work.

“The agentic era demands a fundamentally different approach to infrastructure, one that keeps pace with workloads that reason continuously, scale unpredictably, and operate in production around the clock,” said Chen Goldberg, executive VP of product and engineering at CoreWeave. “What separates infrastructure that performs in a lab from infrastructure that performs in production is the depth of engineering underneath it.”

As enterprises adopt agentic AI, CoreWeave’s milestone with Nvidia puts it at the center of the conversation around scalable AI infrastructure and compute requirements. The June 30 event will address where the neocloud and AI infrastructure markets are headed next

“CoreWeave’s early deployment of Vera Rubin NVL72 reinforces its position as a leading ‘neocloud’ provider focused on rapidly commercializing cutting-edge AI infrastructure,” Nashawaty said. “Systems are actively optimized for the industry’s shift from training toward large-scale inference and reasoning workloads.”

TheCUBE event livestream

Don’t miss theCUBE’s coverage of the “Scaling the Agentic Era” event on June 30. You can also access theCUBE’s exclusive content on demand after the event broadcast.

How to watch theCUBE interviews

There are several ways to follow theCUBE’s coverage of the “Scaling the Agentic Era” event, including theCUBE’s dedicated website and YouTube channel. You can also find coverage from theCUBE events on SiliconANGLE.

TheCUBE podcasts

SiliconANGLE’s “theCUBE Pod” is available on Apple Podcasts, Spotify and YouTube, making it easy to follow the program on the go. During each podcast, SiliconANGLE’s John Furrier and Dave Vellante unpack the biggest trends in enterprise tech — from AI and cloud to regulation and workplace culture — with exclusive context and analysis.

SiliconANGLE also produces our weekly “Breaking Analysis” program, where Dave Vellante examines the top stories in enterprise tech, combining insights from theCUBE with spending data from Enterprise Technology Research. The program is available on Apple Podcasts, Spotify and YouTube.

Guests

During theCUBE’s coverage of the “Scaling the Agentic Era” event, executives from CoreWeave, Nvidia and Dell will talk with theCUBE about the infrastructure, engineering and operational requirements needed to support production-scale inference and agentic AI workloads.

(* Disclosure: TheCUBE is a paid media partner for the “Scaling the Agentic Era With Nvidia Vera Rubin NVL72 on CoreWeave Cloud” event. Neither CoreWeave, the sponsor of theCUBE’s event coverage, nor other sponsors have editorial control over content on theCUBE or SiliconANGLE.)

Image: SiliconANGLE