orange

the July 10 SK Hynix Nasdaq listing is the memory catalyst nobody's talking about. $29B raise, biggest ADR ever, and they own about 60% of the HBM market. i play it through the 2x SK Hynix ETF plus $Roundhill Memory ETF (DRAM.US)$ for the basket. leverage cuts both ways so i keep it small, but the setup into the 10th is too clean to ignore.

Already moved on from Meituan months ago, but still follow the business because the model is fascinating. Core local commerce losses hurt the thesis. Delivery infrastructure moat is real. Just not my risk/return right now 💤

Kuaishou Technology: Can the No.2 Short Video Platform Narrow the Gap?Q1 2026 results test three things: (1) ad revenue sustaining above 20% YoY growth vs. Douyin's dominance; (2) live commerce GMV trajectory as platforms battle for merchant wallet share; (3) overseas traction in Brazil and Southeast Asia. Kuaishou's discount to domestic peers remains wide — a consistent earnings beat could finally close it.

careful conflating SanDisk with Micron though. MU is mostly DRAM and HBM, thats where the AI money is. SNDK is NAND, the supercycle there is real but lags and the pricing is choppier. i hold it but with smaller size than my DRAM exposure, risk-reward isnt the same.

two Singtel moves at once: the DISG AI partnership and the KKR-backed ST Telemedia data-centre deal at over US$3.9B. data centres are the picks-and-shovels of the AI build-out, and Singtel is leaning in hard while still paying a 4%+ yield 🧠

Fed officials have so far resisted characterising the rise in long-end yields as a policy concern, attributing it to term premium re-pricing rather than a signal of tightening financial conditions. But with the 30-year now touching levels last seen in 2007, some officials are quietly monitoring whether higher borrowing costs begin to surface in mortgage delinquency or corporate refinancing data over the coming months. The door to a resumption of rate cuts remains open — the data will have to close it. 🔒

Chip sector mixed—$Intel(INTC.US)  weak, $AMD(AMD.US)  stable—does that last?

down 5% but 42% above IPO and it literally passed Amazon for a hot minute this week. a pullback the day after options list is the most normal thing ever

AVGO down 12.9% AH on a beat AND record guidance 😮‍💨 not selling a single share. $16B in AI chips up 200% is not a business in trouble, it is a stock that ran too hot. I just hold and let the noise clear 💤

Dow up 1.7% to a record, Nasdaq down, semis off 2.1%, Blackstone up 7.5%. This is not risk-off, it is rotation. Money is leaving crowded AI growth and going into value and financials. The index hides it, the internals show it clearly 🧠

I've held COST for years and every time I consider trimming, a quarter like this reminds me why I don't. Q3 FY2026: net sales USD 69.15 billion, up 11.6%. Comparable sales up 9.8%, or 6.6% excluding gas and FX. E-commerce comparable sales up 21.5%. EPS hit $4.93, up from $4.28 a year ago. Every core metric moved in the right direction. The number I pay closest attention to in any Costco quarter is membership renewal. US and Canada came in at 92.2%. Worldwide: 89.7%. 148.5 million total cardholders. These are not one-time shoppers. They're recurring revenue that Costco earns before selling a single item off the shelf. What I find especially interesting is the executive membership tier: 41.2 million paid members, up 9.6% year over year. Executive members pay more, buy more, and renew at higher rates than standard members. As this cohort grows as a share of the total base, membership fee income compounds structurally. That's not a marketing trick; that's a flywheel. COST is not a high-growth stock in the traditional sense. It's a business with genuine pricing power, extraordinary customer loyalty, and a membership model that makes revenue almost predictable. I added during the pullback earlier this year. I'm not touching it.

Let's be real about what the oil market actually is right now.Not OPEC. Not inventory data. Not refinery margins. One phone. Trump posts "deal largely negotiated" and WTI falls 8%. He says "not fully worked out" a day later and oil bounces back. Rubio says "good news within hours" and traders scramble for position. That is your energy market in 2026.If you are holding XOM or CVX and you are not refreshing Truth Social every hour, you are playing a completely different game from the market. I honestly don't know if that makes you smarter or just behind. But that is where we are. 😅

YZJ using internal cash for US$1.05B Poseidon stake = no debt, no dilution. Poseidon becomes an associated company, contributing equity earnings. The deal links YZJ to Fairfax Financial's shipping and energy infrastructure network. For a shipbuilder with strong order books, this is capital recycling into complementary assets. Question is Poseidon's earnings contribution versus the cash drag from now-deployed capital.