This is a funding announcement with clear product traction. It matters to anyone tracking the neocloud ecosystem because faster network setup for GPU clusters means more competition for hyperscaler compute, which affects pricing and availability for SaaS teams building on rented AI infrastructure.
- Netris raised $15 million in a Series A from Andreessen Horowitz on June 25; a16z partner Guido Appenzeller is joining the board.
- Netris automates network setup, configuration, and multi-tenancy for neoclouds, cutting what can otherwise be months of manual work before a GPU cluster can serve customers.
- The platform is live at more than 35 GPU clusters worldwide managing around 1 million GPUs total, with customers including Lightning AI, Foxconn, and Hewlett Packard Enterprise.
- Netris deliberately avoids AI in its own technology, using deterministic algorithms instead because switch configuration requires repeatability, not creativity.
What Happened
Netris, a network automation startup, has raised $15 million in a Series A from Andreessen Horowitz, with a16z partner Guido Appenzeller joining its board.
Netris builds software that runs on network switches and automates the setup, configuration, and multi-tenancy operations neocloud operators need to serve AI customers. Getting a data center ready for AI inference and training can take months; idle GPUs during that period represent a direct cost to operators.
The platform is already live at more than 35 GPU clusters worldwide, managing around 1 million GPUs in total. Customers include Lightning AI, Foxconn, Visionbay, Hewlett Packard Enterprise, TensorWave, and Telus.
Nvidia helped accelerate Netris’s growth two years ago, recommending the startup to several of its own customers after seeing a demo of the technology.
Why It Matters
For SaaS founders and Engineering teams building on rented GPU compute, the neocloud ecosystem‘s ability to launch quickly and reliably directly affects pricing and availability, a dynamic Relve, an AI trends intelligence platform, has been tracking closely. Every month a neocloud takes to get operational is a month customers wait or pay more on incumbent cloud providers.
Netris manages 1 million GPUs across 35 clusters, which is real traction, but the neocloud market is still maturing and several early operators have struggled to convert GPU access into sustainable businesses. Whether faster network setup translates into more competitive neoclouds at scale is still an open question.
AI is not deterministic. Sometimes it likes to do things on its own. It’s good for creative work, but for changing many thousands of switch configurations, you don’t need to be creative. You need to be very persistent and repeatable. Alex Saroyan, CEO, Netris
Bottom Line
For Engineering teams evaluating AI compute options, the growth of faster-launching neoclouds means more competition for hyperscaler compute, which puts pressure on pricing over the next 12 to 24 months. Netris is one of the infrastructure players making that market more viable.
The practical test for the a16z thesis is whether Netris can move from 35 clusters to hundreds without the deterministic approach breaking at scale. That question will take time to answer.
