I started writing about semiconductors on Substack with a thesis nobody was talking about: memory bandwidth, not compute, is AI’s real bottleneck. I’ve been long NVIDIA since 2016. Believed in the vision before it was consensus. Now I’m at GTC with a creator pass interviewing their engineers and covering the architecture shifts shaping the next decade of AI infrastructure. Institutional investors read my work. International press cites it. Raja Koduri quote-tweeted my Intel thesis. I’m an EE, not a finance guy. I just wrote what I saw and NVIDIA kept proving me right. If you have real domain expertise — put it out there. The world needs it. Live from GTC all week.