Foley & Lardner LLP represented TensorWave, an AMD‑backed cloud and AI-hyperscaler running next gen AMD accelerators for high performance AI workloads, in connection with the expansions of space and power at two TECfusions data center facilities.
The deployments add 20 megawatts of high‑density capacity split between Pennsylvania and Arizona, including a new 10 MW footprint at the Keystone Connect campus near Pittsburgh and a 10 MW expansion at TensorWave’s existing Tucson installation, already the largest AMD-based AI deployment in North America.
This new phase significantly expands TensorWave’s capacity and replicates its high‑density reference design across both locations. These expansions extend TensorWave’s established AMD training cluster with additional capacity built on direct liquid cooling and high‑rack‑density designs tailored for large‑scale AI workloads. Scheduled for delivery in the first half of 2026, the projects strengthen a trusted collaboration that continues to set new standards for AI readiness, deployment speed, and sustainable infrastructure.
TensorWave is a cutting-edge cloud platform designed specifically for AI workloads. From training to production, TensorWave delivers unmatched performance with AMD MI300X, MI325X, and MI355X accelerators combined with fully managed infrastructure that scales with customer ambition.
The Foley deal team was led by partner Daniel Farris.
