Fleet adds NVIDIA AI workstations to device management

Fleet has added support for NVIDIA's AI hardware range to its device management platform, bringing NVIDIA RTX PRO workstations and NVIDIA DGX Station systems into the same management environment as existing macOS, Windows and Linux devices.

The update allows IT and security teams to oversee NVIDIA-powered systems alongside other devices already deployed across a company estate. The platform provides visibility into operating systems, hardware, software, packages, security controls and vulnerabilities on those machines.

The announcement reflects a broader shift in how organisations are treating specialised AI hardware. Systems once associated mainly with data centres are increasingly appearing on desks and in standard office environments, putting them under the remit of mainstream IT operations rather than separate research or infrastructure teams.

NVIDIA's DGX Station is positioned as a deskside AI supercomputer designed for building and running AI models. According to the companies, it features the NVIDIA GB300 Grace Blackwell Ultra Desktop Superchip and 748GB of coherent memory.

That creates a management question for businesses adopting such systems. If AI workstations sit alongside conventional laptops, desktops and servers, IT departments still need established ways to track software, apply policy, identify vulnerabilities and document changes.

Fleet says its platform addresses that issue by allowing policies, configurations, deployment waves, reports and remediation to be managed through workflows in Git. It says this gives teams an auditable, reversible record of changes rather than forcing them to rely on ad hoc processes for newer classes of hardware.

Unified oversight

Fleet already supports a mix of endpoint and infrastructure types, including Apple and Microsoft devices, desktop Linux and other connected systems. By extending support to NVIDIA's workstation and deskside AI products, it is positioning AI hardware as another endpoint category that can be governed through the same operational tools.

For enterprise buyers, that could reduce the need to create separate management processes as AI systems spread beyond specialist teams. It also reflects concern among security teams that newer compute devices should be subject to the same compliance and governance checks as more traditional corporate assets.

Mike McNeil, Chief Executive Officer and Co-Founder of Fleet, framed the change in those terms.

"The AI workstation is now an enterprise endpoint, plain and simple. IT teams still need to know what's running, what's vulnerable and what's changed. They already have enough on their plates without building a new management stack for every new piece of AI infrastructure. They should be able to manage AI workstations alongside every other device they own, from one place, with the same visibility and controls. That's what this update gives them," said McNeil.

NVIDIA also backed the expansion as part of a wider push to place more AI computing resources directly in the hands of end users and departmental teams rather than restricting them to centralised infrastructure.

"NVIDIA is bringing powerful AI computing directly to developers, creators, engineers and enterprise teams so they can build and run advanced AI workloads locally. Fleet's support for NVIDIA DGX Station and NVIDIA RTX PRO workstations gives organizations a unified way to deploy and manage their AI systems, helping accelerate AI adoption across the enterprise," said Marriott.

IT responsibility

The development points to an operational challenge emerging alongside the AI investment cycle. As companies buy dedicated machines for model development, inferencing and data science work, they must also decide who manages those devices, how they are secured and how changes are approved.

In many organisations, that responsibility is falling to corporate IT and security teams that already manage broad fleets of laptops, mobile devices and servers. That makes interoperability with existing management practices more important than raw hardware specifications alone.

Organisations in 90 countries use Fleet's software to manage millions of devices. The platform covers device management, patching, endpoint governance and vulnerability management across a broad set of systems.

The addition of NVIDIA hardware support also underlines the growing overlap between endpoint management and AI infrastructure governance. A workstation used by an engineer or developer may now contain computing resources once reserved for a server room, but it still requires inventory controls, software oversight and vulnerability tracking.

Fleet's emphasis on reviewable Git-based workflows suggests it sees change management as a differentiator as AI hardware becomes more common in regular business environments. In that model, policy updates and remediation steps can be inspected and reversed, which may appeal to organisations seeking clearer accountability for how sensitive or expensive AI systems are configured.

The practical effect is to place AI workstations and deskside systems within the same administrative framework as more familiar enterprise devices.