Pat Gelsinger, former Intel CEO, discusses his transition to Playground Global as a venture investor focused on hard technology bets. He shares his framework for evaluating startups (technology proof, market fit, leadership team) and his portfolio thesis around what he calls the 'trinity of computing': classical, AI, and quantum systems working together. Key themes include the need for 10,000x improvement in AI inference efficiency, the importance of heterogeneous compute architectures, the future of EUV lithography and next-generation light sources (free electron lasers), resilient networking for large GPU clusters, the role of dataflow machines, and the return of precision computing for scientific workloads. He also discusses US semiconductor industrial policy, energy infrastructure, and the resurgence of interest in hardware engineering.
Sort: