Skip to main content
F

Fluidstack

Fluidstack builds and operates single-tenant GPU infrastructure for AI labs, governments and enterprises, managing over 100,000 GPUs alongside its Atlas OS and Lighthouse systems.

Fluidstack is an AI cloud platform that builds and operates massive-scale, single-tenant GPU infrastructure for AI labs, governments, and enterprises. The company manages more than 100,000 GPUs, providing clusters of H100, H200, B200 and GB200 systems for training and deploying AI models. Named customers and partners include Anthropic, Mistral, Character.AI, Poolside and Black Forest Labs.

Two in-house systems underpin the platform. Atlas OS is a bare-metal operating system for AI infrastructure, designed to enable fast provisioning and orchestration at scale. Lighthouse is an intelligent monitoring system that optimises workloads and identifies issues before they affect operations. The technical work spans GPU infrastructure, cluster orchestration, high-performance computing, infrastructure monitoring, and hardware, network and storage isolation.

Recent commitments include a $50 billion partnership with Anthropic to build custom data centres across the United States, and a €10 billion investment to build a 1-gigawatt AI supercomputer in France. The company relocated its global headquarters to New York City in 2025 and serves customers globally.

Fluidstack's infrastructure is fully isolated, with no shared resources or noisy neighbours. The company is SOC 2 Type 2 and ISO 27001 certified and GDPR compliant, with a stated focus on enterprise-grade security and compliance.

Open jobs

No open jobs right now.