Hosted.ai is Simplifying AI Infrastructure, Making it More Efficient and Accessible than Ever.
- Karan Bhatia
- Mar 20
- 2 min read
Hosted.ai, helping you turn your GPU infrastructure into a profitable AI cloud, led by Ditlev Bredahl, Narendar Shankar, Julian Chesterfield, and James Withall, has closed a $19M seed funding round led by Creandum, with Repeat VC following, and participation from existing investors People Ventures, Z21 Ventures, Golden Sparrow, Hersir Ventures, and Tekton.
The problem: GPU Infrastructure is Inefficient.
AI relies on GPUs, yet the current infrastructure is rigid and wasteful. Fixed instances based on peak demand create three main issues:
High CAPEX and low profitability: Service providers face massive upfront investments, making profits difficult.
Low utilization: AI workloads use only ~40% of GPU capacity, leaving the rest idle.
Limited access: Regional providers struggle to offer low-latency, local GPU infrastructure due to high costs and scarcity outside hyperscalers.
The Solution: The Hosted·ai Stack
The hosted·ai software stack redefines GPU orchestration, management, and usage:
hosted·ai: Core GPUaaS platform. GPU pooling, multi-tenant workload optimization, and overcommit boost utilization up to 5×, cutting CAPEX, increasing profitability, and opening access for regional providers. GPU Mesh will allow providers to trade spare capacity, enabling neocloud services without extra hardware.
packet·ai: A neocloud powered by optimized customer GPU infrastructure, delivering GPUaaS at competitive prices and supporting an open-source portal for faster market launch.
GPUaaS.com: Wholesale GPU marketplace connecting enterprise buyers with hosted·ai customers and partners for scalable, custom GPU clusters.
Ditlev Bredahl, CEO of Hosted.ai, stated:
“The GPU market faces a waste problem, not scarcity. After 25 years of developing infrastructure software to make service providers competitive, the GPU opportunity is the largest yet. This funding accelerates platform growth, expands partnerships, and extends regional reach. The goal is to build the operating system for the GPU economy, and this round positions us to achieve it.”
