Anthropic Hires Google Energy Leader for Data Centre Push

Anthropic is strengthening its data centre strategy with the appointment of Sana Ouji, a senior energy and infrastructure executive from Google, as it accelerates plans to expand its global footprint.
Sana joins the San Francisco-based AI company after more than six years at Google, where she focused on data centre energy investments and partnerships. Her move reflects Anthropic’s growing emphasis on securing power and infrastructure as demand for AI compute increases.
The hire forms part of a wider recruitment effort that sees Anthropic bring in experienced data centre professionals, many from Google, as it builds internal capability to support large-scale deployments.
Building a dedicated energy function
Anthropic is formalising its approach to energy and infrastructure through the creation of an inaugural energy team. Sana joins alongside Ariel Horowitz and Tim Hughes, both of whom bring experience in grid systems and data centre development.
Ariel, who previously worked in grid modernisation at the US Department of Energy, and Tim, who joined from Stack Infrastructure, add further expertise across energy systems and facility development.
Announcing her move, Sana Ouji describes the transition as a deliberate step. "After 6.5 years at Google, I'm taking on a new challenge," she writes.
"The past few years in particular have been transformational for the energy and data centre industry, and to have had a front row seat to that transition at one of the world's leading AI companies has been a true privilege."
Together, the team reflects a shift towards integrating power strategy directly into infrastructure planning.
A pipeline of hyperscale expertise
Anthropic’s hiring strategy extends beyond energy into broader data centre functions. The company is recruiting extensively from Google, bringing in expertise across design construction operations and security.
Winnie Leung, Head of Data Centre Infrastructure, Brett Rogers, who previously leads data centre construction, and Liwen Mao, Data Centre Design Lead, are among those who have joined from Google. Additional hires include Adam Johnson in electrical engineering and Peter Sarossy in security, alongside operations and system architecture specialists.
This concentration of experience gives Anthropic access to established approaches for building and operating hyperscale facilities. It also reflects the competitive market for skilled professionals in data centre development, where experience at scale is limited.
By assembling a team with direct knowledge of large cloud environments, Anthropic is positioning itself to manage complex deployments that support AI workloads.
Scaling compute and capacity
The recruitment drive comes as Anthropic expands its access to compute infrastructure through partnerships with major technology providers.
The company has secured cloud capacity agreements with Google exceeding 1GW, including access to tensor processing units designed for AI workloads. This arrangement has expanded further, with additional agreements involving Broadcom and Google covering 3.5GW of capacity.
Alongside these partnerships, Anthropic plans to invest US$50bn in US data centres through a collaboration with Fluidstack, with financial backing from Google. These projects are expected to add substantial capacity to support model training and deployment.
The scale of these commitments highlights the link between energy, compute and infrastructure. Data centres must be designed to handle high-density workloads, while also ensuring access to consistent power supply.
Anthropic Chief Executive Officer Dario Amodei has previously acknowledged the complexity of planning infrastructure at this scale, noting that misjudging growth projections or timing could have serious financial consequences.
In this context, the appointment of senior energy and data centre specialists reflects a more structured approach to managing expansion. By aligning energy procurement facility design and operational expertise, Anthropic is working to build a data centre platform capable of supporting its long-term AI ambitions.





