The rapid expansion of artificial intelligence infrastructure has prompted a wave of new entrants into the data center industry. However, launching a data center operation is far more complex than it might appear. Securing the necessary hardware - GPUs, network switches, and storage systems - is only the beginning. Getting all of that equipment configured, operational, and capable of serving diverse customer requirements is an entirely different challenge. For operators aiming to offer cloud computing services tailored to AI training and inference workloads, the setup process can stretch across several months. Every week of delay means expensive GPU hardware sitting dormant, generating costs without producing revenue.

A network automation company called Netris says it has built a solution specifically designed to address this bottleneck for what the industry refers to as neoclouds - smaller, often specialized cloud providers that have emerged to meet surging demand for AI compute. Netris develops software that runs directly on network switches and offers a broader platform that connects to those switches, enabling neocloud operators to dramatically cut the time required to go from hardware installation to a fully operational, customer-ready environment. The platform handles setup and configuration automatically, provides a layer of network abstraction that allows hardware configurations to be adjusted on the fly, and enforces hardware-level isolation between customers - a capability known as multi-tenancy - so that a single cluster can serve multiple clients simultaneously.

A Gap Left by Legacy Infrastructure Giants

The problem Netris is solving is one that established players in the data center industry largely resolved for themselves long ago. Major infrastructure operators such as Equinix, NTT, Digital Realty, Oracle, Microsoft, AWS, and Google addressed network automation and multi-tenancy by building dedicated engineering teams or developing proprietary internal tools. Those resources are simply not available to smaller neocloud operators, who are often trying to get to market quickly without the luxury of large infrastructure budgets or deep technical headcount.

Netris CEO Alex Saroyan explained the technical distinction that sets his company's approach apart from older networking paradigms. Traditional data centers relied on a technology called software-defined networking, or SDN, to manage configuration changes across large numbers of network links. According to Saroyan, that approach is no longer adequate for AI workloads, where data traffic volumes are significantly higher and latency requirements are far more demanding.

"As a GPU cluster operator, you need to make configuration changes to every link, every day. At traditional data centers, they were using something called SDN to do this, but SDN is falling short, because it is a software technology. For AI, software is not okay, because the amount of traffic is so high, everything must be hardware accelerated. So you need something like SDN, but completely hardware accelerated. This is what we do, and this is what we have been doing for eight years."

Saroyan added that the Netris platform is vendor-agnostic, meaning it is compatible with a wide range of networking equipment and industry standards used across data centers. It supports infrastructure built around both Nvidia and AMD server architectures.

Nvidia Endorsement and a Growing Customer Base

The company's approach has attracted notable attention from within the industry. Roughly two years ago, Nvidia reviewed a demonstration of Netris' technology and was sufficiently impressed that it began recommending the startup to its own customers - a significant endorsement given Nvidia's central role in the AI hardware ecosystem. That early validation appears to have translated into substantial real-world adoption. Netris is now deployed across more than 35 GPU clusters worldwide, collectively representing approximately one million GPUs in operation. Customers operating on the Netris platform include Lightning AI, Foxconn, Visionbay, Hewlett Packard Enterprise, TensorWave, Telus, and a number of other operators.

$15 Million Series A Led by Andreessen Horowitz

Building on that commercial traction, Netris has now closed a $15 million Series A funding round led by Andreessen Horowitz, commonly known as a16z. As part of the investment, a16z partner Guido Appenzeller will join Netris' board of directors. Appenzeller brings relevant experience to the role, having spent years focused on enterprise infrastructure and networking technology.

Interestingly, despite operating in the heart of the AI infrastructure sector, Netris itself does not use artificial intelligence to power its platform. Saroyan was direct about this distinction, noting that the company relies on proprietary algorithms developed over the course of its eight-year history rather than AI-driven decision-making. His reasoning is rooted in the nature of the task at hand.

"We started way before AI. We understood the challenge early on, and we started developing this algorithm early on. AI is not deterministic - sometimes it likes to do things on its own. It is good for creative work, but for changing many thousands of switch configurations, you do not need to be creative. You need to be very persistent and repeatable."

The distinction matters in a market where AI is frequently applied as a label to attract investment, regardless of whether it is genuinely relevant to the underlying technology. Netris appears to be taking a deliberate stance in prioritizing reliability and precision over the appeal of AI-powered branding.

Plans for the Funding

Netris intends to deploy the newly raised capital across several areas. The company plans to expand its engineering team to deepen the technical capabilities of its platform, grow its sales organization to reach more neocloud operators, broaden compatibility with additional hardware vendors, and continue developing the algorithmic functionality at the core of its automation system. As the number of GPU clusters worldwide continues to grow in response to sustained AI investment, the market for infrastructure automation tools like those offered by Netris is likely to expand in parallel.

The neocloud segment itself represents a relatively new category within the broader cloud computing landscape - one that has emerged largely as a consequence of the AI boom and the intense demand for GPU-based compute that has followed. For operators in this space trying to compete with larger, more established providers, tools that reduce time-to-market and lower operational complexity could prove to be a meaningful competitive advantage. Netris is positioning itself as a foundational layer for that emerging ecosystem.