chris bee

@Tanr45 @pennycheck Check out @Temple_Eight , great piece on it

@pennycheck Brooo. Never heard of them, sparked mad interest.

@BryzonX Yep, no Numbers what I meant

Some of you may have missed this nugget from the $PENG recent earnings call 🚨 Penguin is now working on PHOTONIC MEMORY Their current KV cache currently uses high speed copper, which gets the job done now however copper can't handle the bandwidth needed for "Agentic AI" without melting or slowing down But as we move towards the agentic ai era, this requires more data, more compute, and more power The only way to scale memory in data centers is with the speed of light If you didn’t know, PENG was an early investor and engineering partner with celestial who are the pioneers in photonic fabric Penguin has been working with Celestial since its early startup days, meaning they have a 24 month head start on understanding how to cool, power, and manage photonic signals One of my fav underrated dynamics of this is Penguin will not need to dilute shareholders to fund this because since Celestial was acquired by $MRVL, Penguin has a war chest of cash from their investment to fund this So what is Penguins role in all of this? Penguin is the one building the Optical Memory Appliance (OMA) the actual physical rack that houses this photonic tech According to their April 2026 roadmap, they are moving toward a commercial launch that will redefine what "memory capacity" means Their current KV cache offers 11TB of memory per cluster, but with their photonic cache they will immediately unlock up to 1000+TB of memory Completely solving the memory problem in data centers (!) $PENG is currently sampling these with "key hyperscalers" ( likely $META and $AMZN ) with final specifications expected by early 2027 If Marvell’s Photonic Fabric becomes the industry standard (which the acquisition suggests it will), every AI data center in the world will need a Penguin built appliance to run it Management noted that their Integrated Memory segment is seeing "favorable pricing dynamics." This is a polite way of saying they are charging a premium for their early photonic and CXL expertise Penguin is leading the way to take the memory wall head on You don’t own enough $PENG

@MegatrendGlobal It rallied before last earnings also, hopefully it will be different this time around

It was great to see $OSS close above $10 and at its weekly high on Friday! That makes me optimistic for earnings week!

@Alex__0x0 Well there was I believe similar situation on IQE (20% up) yet it gapped down 20 proc the next day. Not saying the same will happen

Monday is looking like a major inflection point for Sivers Semiconductors STO: $SIVE / OTC: $SIVEF. While the Stockholm market was closed on May 1st, the U.S. OTC market stayed open and active, creating a massive valuation gap that is hard to ignore. The Monday "Catch-Up" Play The primary listing in Stockholm closed at 38.00 SEK on April 30th. Meanwhile, the U.S. OTC listing $SIVEF surged to close at $4.92 USD on May 1st. •The Math: At current exchange rates, that $4.92 USD close implies a Stockholm price of approximately 45.40 SEK. •The Opportunity: This represents an implied 20%+ gap that the Stockholm market will likely attempt to bridge at the Monday open. Why the Momentum is Real This isn't just about a holiday mismatch; it’s backed by strong fundamental tailwinds in the sectors we've been tracking: •AI Infrastructure Leader: Sivers recently announced a strategic partnership with O-Net and Enablence to develop advanced External Light Sources (ELS) for AI data centers.

 •Massive Market Potential: The Co-Packaged Optics (CPO) market—a key focus for Sivers—is projected to exceed $20 billion by 2036 with a 37% CAGR.

 •The Overlooked SATCOM Engine: Beyond photonics, Sivers is becoming a critical player in SATCOM and defense through its MixComm acquisition, securing its place in the U.S. RF stack alongside giants like Raytheon.

 Final Word With the Stockholm market reopening, all eyes are on whether SIVE will gap up to meet the U.S. valuation. The combination of technical arbitrage and high-growth AI exposure makes this a top-tier asymmetric bet for the week ahead. Financial Disclaimer: I am an independent blogger and investor, not a financial advisor. This post is for informational and entertainment purposes only and does not constitute investment advice. Trading stocks involves significant risk, and you should perform your own due diligence or consult with a professional before making any financial decisions. Past performance is not indicative of future results.

@futurist_lens What about blue moon metals

Rare earths are the choke point. Without them: No AI No EVs No defense No energy Bookmark this. Rare Earth ecosystem: • Rare Earths → $MP $USAR $IDR • Uranium → $CCJ $NXE $UEC $DNN $UUUU $LEU • Lithium → $ALB $SQM $LAC $SGML $RIO • Copper → $FCX $SCCO $TGB $HBM $ERO • Graphite → $NVX $NMG $WWR • Antimony → $UAMY $NVA • Multi Minerals → $CRML $TMRC $UAMY Supply is tight. Demand is locked in. That’s where 10x begins.

@Temple_Eight Bro leave some time for the subscribers to digest before calling in the crowd ;)

Today we take a closer look at Penguin Solutions $PENG This is an extremely undervalued competitor operating in the same space as $SMCI The greatest edge Penguin has is as follows: "In March 2026, Penguin launched the MemoryAI CXL-based KV Cache server, which the company describes as the industry’s first production-ready server built on CXL memory disaggregation architecture. This product directly addresses what engineers call the Memory Wall in large language model inference: the bottleneck that occurs when the KV Cache of a large model running inference exceeds the HBM memory attached to the GPU. The MemoryAI server solves this problem by providing 11 terabytes of CXL-attached memory that the inference workload can address as a seamless extension of the GPU memory pool. The GPU does not need to spill KV Cache to slower storage tiers during inference. The entire context window of even the largest current models fits within the CXL memory pool. The result is inference performance that significantly exceeds what the same GPU configuration can achieve with standard memory configurations." Follow the link in bio or below to read more.

@Ren_aramb Expected more tangible outlook

$LPK / $LPKF earnings dropped and everyone is panicking at the numbers. Revenue down to €17.1M from €25.3M. EBIT margin -40%. . Yet you are looking at the wrong thing. Current financials tell you where the company has been. The earnings call tells you where it’s going. The thesis for $LPKF was never about Q1 2026 revenue. It’s about 2027 LIDE glass core substrate mass production. This is the same mistake people made with $AEHR – selling off on weak financials before the call, missing the entire setup. Good thing I didn’t based my decision only on the financials as the stock ended up almost doubling in price afterwards. What really matters looking into the future is customer pipeline, production timelines, and any indication of high volume orders coming through. Don’t mistake a pre-ramp balance sheet for a broken thesis.

@Temple_Eight Still long based on your research

$MXL. Wasn't finished yet.

@fundmyfund Cut 50 % at 55, rest had a stop loss at 45, so enjoying the ride ;)

Wow $MXL - should have jumped in

@fundmyfund „We are so early”

Hot and heavy discussion on Reddit debating $LPTH drawdown today. 😂 Have to keep reminding myself almost no one knows about this thing.

@fundmyfund Ehh denied once again

Could be the day - lets see how it goes thru the day $LPTH

@Temple_Eight @cesarsroy Would appreciate if let us know when decided ;) here or substack

@chrisbeeSA @cesarsroy Thinking still.

New on sub stack. The CPU Bottleneck: 10 Stocks For The AI Inference Regime. Link below.

@Temple_Eight @cesarsroy Partially or all?

@cesarsroy Thinking of booking tomorrow. I think it can still run to $48 , I'll see how it's behaving in the session tomorrow. If can squeeze above $50 I would definitely book it .

@daniel_koss @pennycheck Guys you should worry about it only after it explodes, it hasn’t happened yet ;)

Every FinX account and their mom is now bullish $OUST Some wrote great articles, but most are honestly generic AI slop copy pasta. Sometimes it's worth mentioning where alpha comes from. The first account I ever saw that mentioned $OUST was @pennycheck over a year ago. And the first account I saw that posted really good takes and insights was @usppdd also over a year ago. It bothers me that now everyone acts like they brought the thesis to X. We should always give credit to who we got alpha from and incentivize sharing alpha on X and punish AI slop. If you share a thesis you saw somewhere else, imho at least add vaue by adding new research or insights you have.

@Temple_Eight Very good content sirs

Thank you to our recent follows & subscribers, we're rising in the ranks!

@fundmyfund If only mutombo could read this!

$LPTH bull case

@fundmyfund Always ahead of the herd

When you claim you bought $ARM yesterday but disclose it today in a serene way; thus already up 10% just like that Speaking of PT Barnum

@daniel_koss Credo

There's a stock crashing hard right now that I believe will be one of the biggest winners in the AI buildout. The narrative says they're dying. My research says they'll print cash for years. I'm not sharing the ticker for now. I'm just hoping it drops more so I can load up on calls 🥰 Will share the ticker and thesis when I take a position though! Just don't want to take any chance on the narrative correcting. Hope more people panic sell all their shares 😂

@stocktalkweekly Let’s all fomo into subscription

LAST DAY! 🚨