Netris Closes $15M Series A Led by Andreessen Horowitz to Accelerate AI Infrastructure Deployments
Network automation startup Netris raised $15M Series A from a16z to help emerging GPU cloud operators reduce time-to-market by automating data center configuration and multi-tenancy management.
Last verified:
Netris Secures $15M to Scale AI Infrastructure Automation
Netris, a network automation startup, has closed a $15 million Series A funding round led by Andreessen Horowitz (a16z), according to TechCrunch. The capital infusion reflects growing investor appetite for infrastructure tools that address bottlenecks in the emerging GPU cloud market, where time-to-deployment directly impacts profitability.
The funding comes as the proliferation of neocloud operators—smaller, independent GPU cloud providers competing with hyperscalers—has exposed a critical gap in infrastructure tooling. While established operators like AWS, Google Cloud, and Microsoft Azure built internal engineering teams and automation frameworks to manage network complexity, newer entrants lack the resources to replicate that work from scratch.
Hardware-Accelerated Networking for GPU Workloads
Netris provides a platform that runs on network switches and abstracts hardware configuration, allowing operators to automate the deployment and multi-tenancy isolation required to serve multiple customers on shared infrastructure. According to Netris CEO Alex Saroyan, the company’s approach diverges fundamentally from traditional software-defined networking (SDN). “For AI, software is not okay, because the amount of traffic is so high, everything must be hardware accelerated,” Saroyan told TechCrunch. Netris has been developing hardware-accelerated networking for eight years and markets itself as vendor-agnostic, compatible with equipment from Nvidia and AMD server providers.
From Nvidia’s Endorsement to 35 Global Clusters
Netris has moved beyond proof-of-concept. Nvidia recommended the company to customers two years ago following an internal demonstration, and Netris now operates across more than 35 GPU clusters worldwide, managing approximately 1 million GPUs collectively. Operators currently running Netris include Lightning AI, Foxconn, Visionbay, Hewlett Packard Enterprise, Tensorwave, and Telus—a mix of systems integrators, telecom incumbents, and specialized GPU cloud startups.
Why This Matters
The proliferation of independent GPU cloud operators creates a long tail of infrastructure operators who must solve data center networking problems without hyperscaler-scale engineering budgets. If Netris can materially reduce the time from acquiring GPUs to offering revenue-generating services—cutting multi-month timelines to weeks—it shifts the unit economics of neocloud startups by reducing idle-hardware carrying costs. This matters most for operators planning to scale from single clusters to multi-region deployments, where network automation becomes increasingly critical to operational complexity. The Series A signals that infrastructure-layer software for AI data centers is attracting tier-one venture capital, potentially signaling further consolidation or exit activity in the GPU cloud tooling space.
Frequently Asked Questions
What problem does Netris solve for GPU cloud operators?
Netris automates network configuration, setup, and multi-tenancy management for data centers, reducing months of manual work to weeks and keeping expensive GPUs from sitting idle during deployment.
How is Netris different from traditional software-defined networking (SDN)?
According to Netris CEO Alex Saroyan, Netris uses hardware-accelerated networking to handle the high-traffic demands of AI workloads, whereas traditional SDN relies on software-only approaches that fall short for GPU clusters.
Who is using Netris today?
Netris operates at more than 35 GPU clusters globally, serving operators including Lightning AI, Foxconn, Visionbay, Hewlett Packard Enterprise, Tensorwave, and Telus, totaling approximately 1 million GPUs in operation.