The race to run AI models at the distributed edge is moving into the real-world environments where enterprise data is created. As more of that execution happens outside centralized cloud environments, edge AI infrastructure is becoming an urgent enterprise concern.

Zededa Inc., an edge computing company focused on helping enterprises deploy and manage applications across distributed environments, is aiming to meet demand with its newly launched Edge Intelligence Platform, designed to bring cloud-like simplicity to AI deployment. The launch reflects a broader inflection point for enterprise edge AI, according to Said Ouissal (pictured, left), co-founder and chief executive officer of Zededa.

“It’s a platform that makes it very easy to deploy agents and models at the edge, building on the same foundation that we’ve already been deploying [for] many customers,” Ouissal told theCUBE. “We’re at an inflection point for inference, and it’s all about now enabling inference anywhere, not just in the cloud, but also in these factories, in these production lines, these stores, these ships.”

Ouissal and Padraig Stapleton (right), senior vice president at Zededa, spoke with theCUBE’s Gemma Allen at the Nvidia GTC AI Conference & Expo, during an exclusive broadcast on theCUBE, SiliconANGLE Media’s livestreaming studio. They discussed edge AI deployment — at scale — across industries, including energy, manufacturing and retail. (*Disclosure below.)

Edge AI infrastructure scales across industries

New hardware is helping drive the buildout of edge AI infrastructure, including Nvidia Corp.’s IGX Thor platform, which brings a level of edge compute capacity that was previously unavailable for industrial deployments. That added capacity is unlocking use cases that were previously out of reach, Stapleton explained.

“With Thor, you can now run a very powerful LLM, VLM model at the edge in an industrial location,” Stapleton said. “You can marry that with agentic AI software and capability, which allows you to run different types of use cases, everything from inspection on widgets going down the line to safety … applications on an oil rig or a platform.”

That broader expansion of edge AI is already showing up across a wide range of enterprise deployments. Zededa is already deployed in more than 100 countries, across verticals ranging from manufacturing and energy to maritime operations with A.P. Møller – Mærsk, Stapleton noted. Another customer is running conversational large language models and computer-vision-driven car identification across hundreds of car wash locations, with plans to scale to 10,000, Ouissal added. Together, those examples point to a market that is moving beyond edge AI proofs of concept and into live operational environments.

“This is well beyond the art of the possible, well beyond what could happen,” Ouissal said. “It is happening, and the winners I think are the ones that go in early and try to amass this technology and put it in their environments.”

Here’s the complete video interview, part of SiliconANGLE’s and theCUBE’s coverage of the Nvidia GTC AI Conference & Expo:

(* Disclosure: Zededa sponsored this segment of theCUBE. Neither Zededa nor other sponsors have editorial control over content on theCUBE or SiliconANGLE.)

Photo: SiliconANGLE