Serenity

Well, looks like SLC NAND is forcasted to rise up 120-170% for H2 2026 per Trendforce. There's $MU (21%), Kioxia (20%) as largest share, but clearer beneficiaries appears to be: 1. Winbond (2344): ~15% of the SLC NAND market 2. Macronix (2337): ~11% market share 3. SkyHigh Memory via Puya Semi: ~14% market share Given Micron at $1T and others are a bit large relative to SLC NAND TAM. (source: Q1 2026 Winbond presentation/Trendforce).

$AEHR back up +36.4% today off earnings! 2027 guide: $130-$150m (160-200% growth) from 2026 revenue. Sees opportunity to guide higher (assumes no memory revenue or little to none from newly benchmarked AI customer). Q4 bookings: $60.7M, effective backlog is $100.6M. - Lead AI processor wafer-level burn-in customer is significantly ramping their products. - Engaged with additional AI processor customers who are evaluating wafer-level burn-in. - Benchmark customer, which was a “major supplier of AI accelerators, CPUs and network processors” “exceeded their expectations. From management: The potential revenue opportunity from one of these …

$NBIS signs $1B+ compute agreement with Reflection AI, for GB300 access through 2029. Reflection also signed a multi-billion dollar agreement with $SPCX earlier. Interesting to say the least, seeing Nebius drop -5% off the news today. Also... counterparty to get this done kinda reminds me of OpenAI, where they might not have the funds to actually execute on these LTAs yet compared to $META or $MSFT. But generally positive long term developments, customer diversification was one of the core strengths of Nebius.

Two different capacity expansion announcements today across $TSEM and $AAOI. Tower Semi, supported by the Japanese government, announced expansion of its 300mm Silicon Photonics (SiPho), Silicon Germanium (SiGe), and advanced packaging capabilities. Targeting $1.2B net profit, and $3.6B revenue in 2028 (with a ~$28B MC, that's ~23.3 2028 forward p/e). AAOI announced the commencement of its buildout, roughly 400,000 square feet of manufacturing to make 800g/1.6T production go brrr. The overall expansion is also supported by the local city government in Texas. (they also $20.85M worth of subsidies from the State earlier). This one isn't ex…

So Jensen himself went out to do damage control with $NVDA delays reports apparently: 1. Jensen Huang said no delays: - 800V and rack-to-rack optical interconnects is progressing fully in line with the original plan with no material delay. - Debunked the rumor that the next-generation flagship, Rubin Ultra, would be delayed until 2028 and said it would be on track to ship next year - Rack system optimizations, like replacing Kyber with a new design are architectural optimizations, not impacting delivery timelines. 2. On ASIC competition: no loss of market share. - Talked about how a leading frontier lab that previously relied heavily on…

TLDR of Innolight investor relations takeaways: 1. "Overall, 1.6T market demand has not contracted; instead, 800G demand has increased significantly compared with previous expectations" Prob most important takeaway as a whole was 800G demand revision (also longer tail demand). Which is a bullish read through on US transceiver makers like $AAOI, $COHR, $LITE. Lot of new customers like neoclouds, AI model companies, contributing to overall demand rather than just hyperscalers, diversification always a bonus. 2. Innolight said the shortage covers the module supply chain broadly including: - Optical chips. - Electrical chips. - PCBs. - Other…

Morgan Stanley note on CPO today. Key participants include: - $SIVE, $COHR, and $LITE in laser supply - Broadcom and Nvidia in switch platforms - Lightmatter, Ayar Labs, Marvell/Celestial, and POET in optical engines and photonics, - $TSM, $GFS, and $TSEM in silicon photonics foundry capacity I’ve covered all of these before, like Tower Semi. But I’m especially happy that Morgan Stanley validated my research that $SIVE is one of the critical global players in CPO. A small $1.5B laser company next to your two leading $60B+ companies… If you also synthesize Rosenblatt Securities recent note that China laser suppliers were quite far from ha…

Just a semi recap TLDR: - $GLW glass bridge per Morgan Stanley has potential, but hard to displace FAU (like FOCI) in short term - $SPCX Starlink Gen 3 is scaling to 100,000 units (10x prev gen) creating possible capacity constraints for suppliers of switches to CCL. - PCB supply shortages are projected to persist until 2028, and component shortages/price hikes are already forcing ODMs like Inventec to issue conservative H2 shipment - DeepSeek and Zhipu are developing custom ASICs to bypass $NVDA (kinda expected by now). - Anthropic has achieved a $30B ARR and is projected to hit >$1B in Q3 profit. Turns out these frontier labs are more …

Looks like the Strait of Hormuz is open for the 50th time this year per Trump. I’m getting tired looking at macro. https://t.co/GuYgkwC2Q8

So there's a IBK Research report on Boston Dynamics value chains from last month. Just a summary: IBK maps these companies to Atlas as suppliers: - Hwashin (010690) / body, arms, legs - LG Energy (373220) / battery - Hyundai Autoever (307950) / integration - Hyundai Mobis (012330) / actuators As for humanoid volume ramps: They're modeling for, 11.29K in 2028, 20k 2029, 30k in 2030... 40k in 2031, and 50k in 2032. Not quite sure why IBK and other institutions are a fan of linearly modeling S-curve volume ramps... Like adding +10K per year, don't quite think it's volume ramp is going to work like that... if I had to guess it would look mo…

Apparently, Korean Beauty was where all the alpha is. APR (278470.KS) is up 529% over the past 2Y, wildly outperforming both $NVDA and $INTC. Every female I know talks about their "Medicube" stick, then buying products... where you put a whole salmon's family tree on your face for better skin. The world's female population creates a bigger structural demand than hyperscalers for memory or lasers? Should have known...

Contrarian take, but Sam Altman / OpenAI is doing something right if he's pissing off every Mag7 like $AAPL. By building their own ecosystem instead of being sucked into another. It's been awhile since $GOOGL, Apple, Microsoft, and others had some genuine disruptors.

Regarding $AMZN RNG + OpenAI’s MRC and VCSELs. In what appears to be an indirect response on B. Riley's $AAOI sell report regarding optimizations that "flattens networks and cuts transceiver TAM 40-50%". Rosenblatt TLDR: They've already modeled for this, even though each XPU may use fewer transceivers. The number of XPUs is growing so quickly that total optical demand should still rise. On lasers architectures for CPO: VCSELs are useful as a credible bridge for NPO/short reach. Ideal architecture though for CPO is UHP CW over other lasers, and cites Broadcom for that statements. Which happens to supports my core $SIVE thesis. I feel lik…

Rosenblatt on recent optical sector weakness from $AAOI to $LITE: "Stocks in the Optical sector have traded poorly for the last 1-2 months" due to CPO delay reports and China capacity scares. "We think short sellers... do not actually strongly believe in the thesis." Multiple short sellers told them they will likely close their positions... late July and early August. Further, they say they plan to buy Optical stocks in 2027 for the scale up CPO opportunity. Hilarious report on laser weakness, read is very positive for $SIVEF, $COHR, and others moving forward. I've been long on CPO sector, but seems like institutions want entry points, j…

Sk Hynix is now trading under $SKHYV! It's now up an insane 16%. I wonder how Sk Hynix over in Korea is going to react on Monday. https://t.co/Mmqdu0poVY

Mitsubishi Motors joins robotics race to deploy humanoid workers by 2027. Pairing with Highlanders, a Japanese robotics company. - Agility $CCXI x Foxconn, Toyota - Apptronik x $JBL, Mercedes - Highlanders x Mitsubishi - Figure × BMW - Boston Dynamics x Hyundai (subsidiaries) - Optimus x Tesla (Internal) - Rainbow Robotics x Samsung (subsidiaries) Idk if it's just me of there's a lot of patterns with humanoid companies playing buddy-buddy up with a major auto player/manufacturer. If they're not already vertically integrated by one already. Then the Chinese players like UBTech/AgiBot/Unitree don't seem to care and work with a wide variety…

The "Mythical Golden Egg Dragon Candle" predicted the $SIVE rally today. If you look at the Golden Egg hatching indicator I drew, you can tell it predicts Sivers CEO buying on the open market a day later. Do you guys think it's time I should sell courses on TA? https://t.co/LweBzWStuk

Yep, it’s interesting to witness retail capitulation. After we got Bloomberg Meta compute article cleared up by their internal memo. And the 2 reports on delays were denied by Nvidia. Both of which caused the selloff in the first place... So Photonics / Nvidia’s roadmap hasn’t fundamentally changed: $LITE is still completely sold out for the next 2 years and likely into 2029. $SIVE is about to volume ramp with GlobalFoundries, Jabil, Poet, Ayar, and other hyperscaler suppliers. $TSM COUPE and their TW suppliers from Shunsin to Foci aren’t randomly disappearing into the void. $AAOI and their $471m/month h2 2027 revenue projections have…

"China is shipping ~90% of the world's humanoid robots. We cannot keep taking them lightly." Elon's comment: "For now" Meanwhile, 1X today posts a video of the most advanced humanoid hand. Looks like the inflection point for American humanoid programs is here? Both technologically with players like 1X and with mass production with $TSLA? I'm am personally long on Agility through $CCXI targeting commercialization with Amazon and others. But I'm proud to see American robotics as a whole go brrrr.

Three new $SIVE insider purchases today. Most notably: $SIVEF CEO Vikram bought 1,000,000 SEK worth of shares on the open market today. This is following the massive drop on Sivers stock following 2 misleading CPO delay reports, Meta Compute headlines, and another insider sale report from another photonics director. Typically, insider sells for any number of reasons. But having the CEO buy on the open-market is the clearest signal in management confidence. Following $GFS, $JBL, and other laser supplier partnerships, and his goal of NASDAQ listing in the near term. Feels like the CEO sees a clear potential for Sivers to become the next ph…

Nothing like an internal $META memo getting published. Showing AI ramping as usual: - LTAs signed with Samsung and $SNDK for memory - LTAs signed with Sumitomo Electric for fiber optics - Expected to deploy 7GW compute infra this year, and doubling in 2027. - as much as $145B capex spend this year “It plans to launch a chip about every six months through 2027”

Wow, looks like Elon + xAI at $SPCX managed to build an Anthropic Opus-level LLM with Grok 4.5. That’s much more cost efficient as well… Kinda emulating what Chinese vendors have been doing to compete on pricing + adoption. (Which is an ideal thing given Chinese model growth among US companies recently) Never doubt Elon, maybe they might end up leapfrogging Anthropic + Mythos?

$META to build a new $10B DC in Canada to expand AI capacity. So much for the media framing of “Meta Compute”as overbuilding and cutting capex. https://t.co/0QgSls0s9o

Let’s put it this way: Retail thinks a 30-60% drop is a “falling knife” where bagholders will never recover. And they end up panic selling after seeing swiggly line TAs and people comparing valuable chokepoints to memes. I see it as long term ownership over: - the next hyperscaler with $NBIS, projecting $7-9B ARR Q4 - 40% of the InP supply chain with $AXTI - leader of the next optical shift with $SIVE with CW DFB lasers - leader of US humanoids with $CCXI inside a future trillion dollar theme And so on… At cheaper valuations. While I might do things like lowering margin or hedging with my own portfolio, I’m personally not panicking when…

Blue Origin reportedly raising $10B at a $130B valuation per NYT. Seems pretty positive for $RKLB, $ASTS, and other space names in terms of relative valuation anchoring. If SpaceX and Blue Origin is able to set such high valuations. https://t.co/04W7phuIwr

@zjcwebster My prediction is we’re going to cut off trade with Belgium next. There goes all the chocolates. https://t.co/y4CDTx5ZcT

@AlienMShroom Spain has lithium, tungsten, and other rare earths America needs. That China happens to be export controlling…The US needs all the partners it can keep.

Trump: “Spain is a terrible partner in NATO. They don't participate. They don't pay. Cut off all trade with Spain” Tbh no clue what’s going on anymore… https://t.co/F1n0m6gcPa

I mean I’m pretty confident memory and AI names recover just cause of the whopping amount of operating profit/growth they have. I don’t think there’s anything fundamentally wrong, especially after the Samsung earnings. Just a pretty wild deleveraging/margin cascade right now across the board with many things crashing 30-60% already. Short term option traders and 2.5x margin bros were probably wiped.

@zephyr_z9 Koreans already fully liquidated a week in. https://t.co/DNIx7q2kW8