VP of Engineering on AI Infrastructure Roadmap and GPU Procurement Strategy at NVIDIA

This 52-minute expert call features a current VP of Engineering at NVIDIA discussing the company's AI infrastructure roadmap through 2027. Key topics include the internal GPU procurement pipeline and allocation strategy between cloud hyperscalers and enterprise customers, competitive dynamics with AMD's MI300 series and Google's custom TPU chips, pricing power sustainability as inference workloads scale, the transition from training-dominated to inference-dominated GPU demand, and NVIDIA's software moat through CUDA ecosystem lock-in. The expert provides firsthand perspective on supply chain constraints, margin trajectory, and strategic priorities that are not yet reflected in sell-side coverage.