OpenAI Appoints Former xAI Executive to Lead Compute and Infrastructure
Uday Ruddarraju assumes control of OpenAI's massive data center and compute operations strategy.
1Leadership Consolidation Strategy
OpenAI has officially merged its compute and infrastructure divisions under the leadership of Uday Ruddarraju. Ruddarraju, previously an infrastructure executive at Elon Musk’s xAI, assumes these combined responsibilities immediately. This structural change follows the departure of Sachin Katti, who previously oversaw infrastructure and led the significant 30 billion dollar Georgia facility project.
2Transition of Key Personnel
Sachin Katti, who joined from Intel late last year, will step down from his primary role within weeks. Despite his departure from the executive position, Katti will remain with OpenAI as an adviser to ensure a smooth transition. The existing business leaders reporting under Katti are expected to retain their current roles.
3Ruddarraju’s Technical Background
Uday Ruddarraju joined OpenAI in July 2025 and served as the chief technology officer for compute. His experience includes managing large-scale training runs, specifically working on Grok 3 across 100,000 GPUs during his tenure at xAI. This technical background is central to his new mandate of managing OpenAI's massive, capital-intensive compute infrastructure.
4Infrastructure Project Scope
The infrastructure portfolio includes the 30 billion dollar Georgia site, which was the first facility OpenAI designed and led independently. Additionally, Katti oversaw an 8-gigawatt project in Ohio developed alongside SB Energy and Nvidia. Nvidia has reportedly guaranteed up to 105 billion dollars in lease and power payments for this specific Ohio initiative.
5GPT-Infra Division Realignment
The GPT-Infra unit, which utilizes OpenAI models for semiconductor design, is shifting under the leadership of chip chief Richard Ho. This transition occurs as Ho has expressed reservations regarding the reliability of model-generated design changes. The move aims to better integrate chip development efforts with the company’s broader hardware and infrastructure strategy.
6Geographic Focus and Constraints
OpenAI’s infrastructure strategy is heavily concentrated in the United States. The company paused its Stargate UK project in April due to British electricity prices being four times higher than US levels and unresolved copyright regulations. Consequently, the vast majority of the 856 billion dollar compute investment is allocated to projects across the Atlantic.
7Recent Talent Departures
The leadership change follows a period of personnel turnover within the infrastructure team. Chris Malone exited the company, and Peter Hoeschele moved to Meta earlier this year alongside two colleagues. Data Center Dynamics also identified Keith Heyde and Shamez Hemani as departures, highlighting a broader shift in the technical team composition.
8Future Operational Outlook
The appointment of Ruddarraju signals a shift toward more unified compute management as OpenAI scales its hardware footprint. With the Ohio project dwarfing the 230-megawatt Norway site by 35 times, the company is prioritizing massive domestic capacity. Ruddarraju must now balance these commercial partnerships while navigating the complex demands of large-scale AI infrastructure.