STL Launches Neuralis for Telcos with AI Workloads in the US

STL Optical Connectivity, a subsidiary of STL, has introduced its Neuralis suite in Washington, DC, with a focus that extends beyond data centres into telco networks. 

The Neuralis portfolio is positioned as a connectivity layer for modern digital infrastructure, where network performance depends on how efficiently large volumes of data move between distributed systems.

This includes not only core data centres but also metro networks and interconnect points that support mobile and fixed services.

AI training workloads rely heavily on GPU clusters, which in turn depend on dense, high-speed fibre connections. This requirement feeds into backhaul and fronthaul networks as well as data centre interconnects that carry traffic between facilities.

STL presents Neuralis as a response to these pressures, drawing on customer collaboration to address constraints around space and deployment speed.

With a growing footprint across the US, the company targets operators that are scaling infrastructure to meet AI-driven demand across their networks.

Traffic patterns reshape telco networks

A key theme in the launch is the shift in traffic patterns, which has direct implications for telco architecture.

Traditional models prioritise north-south traffic, where data moves between end users and central servers. AI and hyperscale environments instead generate east-west traffic, where data flows laterally between nodes.

This change increases the importance of high-capacity interconnects within the network for telco operators.

Data is no longer simply delivered to and from users, as it is processed, analysed and redistributed across multiple locations. This places greater strain on fibre networks, particularly in metro and core segments.

GPU-heavy systems require thousands of fibre connections to maintain throughput, while also operating within power and space constraints.

Neuralis addresses this by introducing high-density fibre cabling that supports large-scale interconnections without expanding the physical footprint, which is useful for telco sites with limited space.

The suite is structured around two main areas. The first focuses on optimising usable space through pre-terminated fibre cables, which are assembled and tested before deployment. This reduces the need for on-site installation work and helps accelerate network rollouts, particularly in dense urban environments.

The second area centres on data centre interconnect (DCI). This underpins the links between core sites, enabling traffic to move between network nodes at scale. 

High-density fibre for network capacity

Within the DCI category, STL’s Celesta IBR series is designed for high-density deployments that align with telco requirements for capacity and efficiency.

The cables support up to 6,912 fibres within a compact form factor, allowing operators to increase throughput without expanding duct space or cable infrastructure. Increasing fibre counts within existing infrastructure helps operators avoid costly physical expansion while meeting rising demand.

The design also takes into account thermal and safety considerations. Dense hardware and high-capacity links generate additional heat, which must be managed to maintain network stability. This is a factor in both data centre facilities and network aggregation sites.

Ankit Agarwal, Managing Director, STL says: "AI demands a level of precision and density that traditional cabling simply cannot meet.

"With STL Neuralis, we are providing the high-speed, low-latency foundation that allows GPU clusters to perform at their peak, moving complexity out of the field and into a controlled, high-precision factory environment."

STL’s vertically integrated manufacturing model covers the full lifecycle of fibre production. It begins with ultra-pure preform, where silica glass is prepared, followed by fibre drawing, where the glass is formed into strands with precise characteristics.

These fibres are then assembled into cables and connectorised into plug-and-play systems.

Deployment and operational efficiency

The launch also highlights a shift in how telco infrastructure is deployed. By moving assembly and testing into factory environments, STL reduces the complexity of field installation, which lowers deployment risk and improves consistency across network builds.

As networks expand to support AI and data-intensive services, the ability to deploy high-capacity connectivity efficiently is a core necessity.

Neuralis positions fibre infrastructure as a core component in addressing that challenge, linking data centre performance with the broader demands of telco networks.