🟢 [E1660] Portfolio construction: Long atoms and long Bittensor. Both can win—if centralized hyperscaler investments are justified, atoms outperform; if Chinese FOSS model wins, Bittensor emerges.