The $7B Startup Building AI Chips to Challenge Nvidia | Fractile Founder & CEO, Walter Goodwin
Walter Goodwin, founder and CEO of AI chipmaker Fractile, breaks down how memory bandwidth bottlenecks limit AI model inference and why 3D stacking DRAM atop logic dies provides a superior architectural alternative to SRAM or standard HBM chips. Goodwin also discusses Fractile's decision to bypass Broadcom, the geopolitical energy crunch facing Western data centers compared to China, and why chip heterogeneity risks creating stranded capacity.
- Hyperscalers and startups alike, including Google for TPU, Meta for MTIA (Jalapeno), MatX, and Etched, rely on Broadcom to handle backend physical design and foundry execution. Read →
- The primary ceiling on AI expansion has shifted from GPU allocation and wafer supply to grid capacity, physical land, and substation hookups. Read →
- Neocloud GPU hosting is commoditizing rapidly because competent hosting providers offer almost zero technical differentiation once clusters are online. Read →
- Data centers that pair GPUs like Nvidia Vera Rubin with SRAM chips like Groq face severe utilization mismatches when AI workloads shift. Read →
- Model training is shifting into a fixed upfront cost, making inference and test-time compute the primary operational expense of frontier intelligence. Read →
- Compute power scaled 1,000,000-fold over the last 25 years, while memory bandwidth improved only 40-fold, creating a 25,000-fold bottleneck. Read →