@yasutaketin posted 237 times across 80 distinct tickers over the past 12 months, accumulating 401 cashtag mentions. The voice concentrated activity heavily on semiconductor and AI infrastructure plays, with $NVDA dominating at 86 mentions, followed by $GOOGL at 36 mentions and $MU at 35 mentions. Most recent activity centered on comparative analysis of chip architectures and their implications for AI model development.

On June 4-5, 2026, @yasutaketin published detailed technical analysis comparing FP4 floating-point implementation across competing chip architectures from NVIDIA, Google, and Amazon. The posts outlined performance differentials in AI training infrastructure, discussing how architectural choices in newer generation chips ($NVDA GB200/VR200, $GOOGL TPU 8t, $AMZN Trainium 3) impact computational capability for large language model development. The analysis included quantitative comparisons of FLOPS performance metrics and datacentre-scale deployment scenarios through 2027.