Matthew Murphy

Background

About Matthew Murphy

I am passionate about helping businesses harness the full power of AI through our cutting-edge cloud platform, tailored for large-scale machine learning operations. At Nebius, we excel by offering multihost training on thousands of H200/H100 GPUs, all interconnected via the latest InfiniBand network, delivering unparalleled speeds of up to 3.2Tb/s per host. Our platform not only provides significant cost savings—with GPU compute costs at least 50% lower than those of major public cloud providers—but also ensures seamless integration and optimization through dedicated engineer support. We focus on simplifying the management of ML frameworks, offering fully managed Kubernetes or Slurm for efficient job scheduling. Our marketplace is packed with a variety of ML-focused libraries, applications, frameworks, and tools to streamline the model training process. With comprehensive documentation and multiple resource management options via our cloud console, CLI, or Terraform, Nebius makes the journey from setup to execution effortless. Plus, accessing virtual machines is a breeze with SSH. Nebius is more than just a GPU cloud; it's a complete platform designed to meet all your training needs.