ScaleOps, a startup that automates Kubernetes infrastructure management, announced a $130 million Series C today at a valuation exceeding $800 million. Insight Partners led the round, with Lightspeed Venture Partners, NFX, Glilot Capital Partners, and Picture Capital all participating. Total funding now stands at $210 million.

The Problem

AI deployments run on infrastructure that was designed for predictable workloads. When hundreds of models and agents share a cluster, demand fluctuates constantly — but most engineering teams still configure resource limits manually or in static blocks. The result: idle GPUs sitting at full allocation while other workloads starve.

ScaleOps addresses this by continuously monitoring live performance signals and adjusting CPU, memory, replica counts, and GPU allocations in real time, without human intervention. The company claims its platform can reduce cloud and AI infrastructure costs by up to 80%.

Why Now

The AI infrastructure management category barely existed two years ago. Today it represents one of the clearest cost pressures facing any company running models at scale. Every major cloud provider now charges significant premiums for GPU availability — and enterprises are starting to count idle capacity as a budget line item.

ScaleOps says it will use the funding to expand autonomous management beyond compute to cover the full resource stack, grow into new enterprise markets, and scale its engineering team.

The round reflects a broader pattern: infrastructure tooling that was once treated as overhead is increasingly positioned as a direct lever on AI margins.