Herman Jin

Herman Jin

253 Twitter followers

Follow

Ex-Goldman Sachs Asia FICC Exec & Co-founder of Zen Family Office. Now diving into crypto, exploring blockchain's potential.

Posts

Another Minneapolis resident has died. In times of extreme public opinion, the middle voices will be ignored. Whether it will be an incitement for a general strike or an armed protest remains to be seen. Any high-beta investment should be approached with extreme caution. I only dare to hold large-cap stocks that haven't fallen too much, and even then, I feel very uneasy.

Claude Cowork aims to prove the feasibility of programmable agency agents. I feel like AI is about to be reignited. The main theme of the past few months has been abundant capital and disbelief; the market rushes in and hypes up any perceived weakness or shortage. However, there's a lack of trust in the essence of AI. "Accessories" are being overpriced, while mainstream, profitable technologies are ignored. The debate over narrative and profitability is unsustainable. It's like a pressure cooker that's been simmering for an hour and is about to burst.

Ryan Leung 梁興盛, CFA

2–3 year useful life, they said. But H100 rental prices are climbing again... @michaeljburry Michael Burry must be revisiting his model😎 $NVDA $ORCL $MSFT $AMZN LFG!!🚀🚀🚀🚀 x.com/matthew_sigel/…

"Long slope and thick snow" refers to my strong long-term optimism, but also the need to be clear-headed about short-term profits and margin expectations. Chen Liwu could easily speculate in stocks now; simply reduce production and raise prices, and the margin, profits, and stock price would immediately soar. But is this what long-term investors and the industry expect from him? Lipbu Tan has grand ambitions, buying at such a significant industry bottom. Even if he can't become the next Steve Jobs, he should at least surpass Lisa Su.

别瞎解读，我说下INTC看法 1. 缺货CPU不是18A，而是intel 3/7 2. 产能瓶颈在packaging，虽然技术不高，但非全自动的，产能提高很慢 3. 晶圆厂供不应求之际选择盈利与margin不一定是最优解。可能更好的途径是增产、爬良率，为今后打基础晶圆厂非一蹴而就，不能既要还要 INTC长坡厚雪 x.com/kayliatyyy/sta…

Do you still think my selling INTC shares to buy large amounts of MSFT and NV is too naive? Don't always rush into investing; be cautious. If you sell at the peak, you'll regret it. I'm starting to feel that way now. Wafer fabs can only increase margins after they improve capacity and yield. Don't overhype them overnight; the semiconductor industry takes ten years of hard work before a single big win. However, after pouring cold water on this, let's give INTC a boost. "Long runway, deep snow"—in five years, you'll see a different landscape in the semiconductor industry, and INTC will be a crucial part of that.

你可以买任何一支大票：MSFT/AMZN/NV/AAPL，但就现在不能买goog x.com/luckyzzzzzz202…

Orcl allows you to buy options but not go all-in on the underlying stock. Going all-in on NV, MSFT, and AMZN isn't actually that problematic. The key this year is to control drawdowns.

Previously, the groups that attacked me accused me of being bearish on BTC and bullish on US stocks, and of encouraging retail investors to buy INTC. Now, what do they usually accuse me of? That I scammed 5 billion? Or that I'm a womanizer? I've blocked them; I can't see them anymore.

I was a die-hard bull on BTC during the most pessimistic period in the crypto, and my fellow crypto enthusiasts are still criticizing me. Is it so hard for me to admit that I'm weak?

估计看不到韭菜割大户的戏码了在币圈，散户们为什么总是输的一方。美股去年散户表现都要好于基金经理的 x.com/shanghaojin/st…

Quietly wait and see if the hype surrounding CPUs starts. Sometimes, knowing too early can lead to losses. In a few days, the "talking heads" will officially start hyping it up. Please follow my rules: Either don't believe it, or believe it early; If you don't believe it at the beginning, absolutely don't believe it. By the end of last year, everyone in the industry knew about it. Only after everyone had built up their positions did the "buying wingmen" emerge. I want to ask: Ridiculous! Why believe everything New Eastar says, but not Nvidia/INTC's financial reports? The Q4 report clearly stated a CPU shortage.

看到存储涨价后MU WDC SNDK的走势了吗？如果我告诉你明年CPU涨价50%，涨了还缺货呢？

I don't know if you understand this: TSM was the best "gatekeeper" for the AI bubble. Figure 1: TSMC CoWoS Figure 2: TSMC Capex

It's not really about how much you lose if SNDK doubles, because the process is random. If it rises to 450 in a month, who would dare hold a naked put position? It's the same principle as short Altcoin; if someone pumps the price, you short at the peak. Those who can hold on are the true heroes (and often become martyrs).

还成吧，420有$11k一手，令人震惊。没涨到532 都是赚的

Loading..