Ze Dude

💯 Totally agree @athuinvests. Institutions will likely short after the Nasdaq listing to create shakeouts and scare retail out. The key is not to fall for it. Stay focused on the long-term thesis: WIN scaling, LITE/Coherent fully booked through 2027-28, Ayar/Jabil/O-Net progressing, and a solid Wireless/Defense leg accelerating (CHIPS Act + YSS/ALL.SPACE). Strong community = we hold through the noise 💪

1) It is likely to be shorted by the institutions. We just need to make sure enough of us are prepared for that and spread the word. That's why the community building is happening. 2. I think dilution will happen with Sivers for accelerating their growth. When? Difficult to say, because that is totally up to them. But one thing I am sure is, just like $AAOI did the same to expand manufacturing capacity, $SIVE could do that too.

Two $SIVE questions: 1) NASDAQ listing is great for availability to institutions. But isn’t it then available for them to short it too trying to force retail out? 2) Separately at what point does dilution become a possibility? @mkfilko @athuinvests @PepInvestStocks @Ren_aramb

@PhotonCap Why? Can you tell us more or link to one of your post if you already answered ? Thx 🙏

이 대답이 100% 옳다곤 생각 안함

#AIOptics #SiliconPhotonics #SIVERS #LITE #AI #Photonics #DataCenters 🚨 Goldman Sachs ultra-bullish on AI Optics! New report: optical networking TAM explodes from $15B to $154B (10x) by 2027-28 — driven by AI data center scale-up + silicon photonics penetration jumping from 6% → 46%. Plays to watch: $LITE $COHR $AAOI $SIVERS Summary 👉 moomoo.com/community/feed… #AIOptics #SiliconPhotonics DYOR — not financial advice

AI CPUs Are Devouring DRAM — Memory 'Shortage' to Last Another Year The memory industry has posted record-breaking results driven by commodity DRAM prices that have surged more than 100%, and with the proliferation of AI-purposed central processing units (CPUs) now coming into play, forecasts suggest the shortage will extend for another year. Intel's recently unveiled "AI CPU" is expected to carry up to four times more commodity DRAM than previous generations. Combined with surging demand from graphics processing units (GPUs) that require high-capacity DRAM, observers expect the memory supply capacity of Samsung Electronics and SK hynix to fall short of demand. According to industry sources on the 2nd, CPU manufacturers are pursuing the integration of 300–400GB of DRAM into AI CPUs. This is up to four times the overwhelming scale compared to typical CPU products (96–256GB). CPUs Emerge as the 'AI Orchestrator' The surge in high-capacity DRAM demand for AI CPUs is tied to the AI industry's pivot toward an inference-centric structure. Whereas AI inference was once limited to simple Q&A, it now serves as the "orchestrator" coordinating various agentic AIs. The key in this process is "context memory." For a CPU to coordinate the entire workflow by referencing the outputs of each agentic AI, it must remember the content. This makes scaling up memory — the space for retention — essential. Until now, AI data centers have built computing infrastructure centered on GPUs equipped with High Bandwidth Memory (HBM). Leveraging the GPU's strength in training AI on vast amounts of data simultaneously, the focus has been on "AI training." Server configurations accordingly followed an 8-GPU-to-1-CPU pattern. However, as the industry's center of gravity shifts toward inference, server configurations with substantially expanded CPU ratios are spreading. In a recent earnings call, Intel executives explained: "In AI inference infrastructure, the computing structure has shifted to a CPU-to-GPU ratio of 1 to 4, and the trend is moving further toward 1 to 1." After GPUs, CPUs Join the Memory Battle — Demand Snowballs The competition for memory capacity has expanded from GPUs to CPUs, snowballing in scale. NVIDIA's next-generation AI chip "Vera Rubin" carries 288GB through 8 HBM stacks, while AMD's next-generation GPU MI400 boasts an even larger 432GB mammoth-class capacity. Google's recently unveiled custom chip — the 8th-generation Tensor Processing Unit (TPU 8i) — is also slated to feature 288GB of HBM. Add to this Intel's AI CPU "Xeon" and AMD's "EPYC" beginning to use up to 400GB of high-capacity DDR5, and the memory shortage is expected to persist longer. The market temperature is already being proven in spot prices. According to Kiwoom Securities, while the price of legacy DDR4 (16GB basis) plunged 16% in a single month in April, the spot price of DDR5 (16GB basis) — used in AI CPUs — rose 2.8% over the same period, maintaining its price premium. An industry source said: "The current DRAM market is understood to be roughly 10 percentage points short of demand. With commodity DRAM demand surging on top of HBM, the supercycle is highly likely to extend from the previously expected 2026 into 2027." $MU $DRAM

#IREN #Sweetwater #AIInfrastructure #PowerIsKing MEGA NEWS for $IREN 🚀 Sweetwater 1 (1.4 GW) has been successfully energized! After months of waiting, patience is finally paying off. CEO@danroberts0101 confirmed it: delivered on schedule, flawless execution, and a vertically integrated model that delivers. Power is still the bottleneck in AI. No power = no GPUs = no AI. IREN doesn’t just build data centers — IREN delivers the power AND the infrastructure. That’s the game-changer. In the AI world, energy is at the center of everything. Whoever controls the energy, controls it all. Next stop: Q3 earnings on May 7. x.com/IREN_Ltd/statu…🚀🚀

@Kaizen_Investor Haha hell yeah it works like magic! SuperGrok just went full Kaizen mode on my entire portfolio in one shot , complete with brutal red flags, dilution roasts, and Say-Do Ratio scores ! Super useful ! Thx again for it mate !

@ZeDude1510 Haha nice, does that work?

The rating of my current holdings according to this framework. $GOOGL: 97/100 Alphabet’s management operates with relentless execution and immense capital discipline. While the sheer, absolute dollar magnitude of their stock-based compensation remains staggering, their aggressive share repurchases, strict strategic focus on monetizing AI infrastructure, and transparent, data-backed communications make them elite operators who are successfully defending one of the largest economic moats in existence. $FTC: 96/100 Filtronic’s management team under CEO Nat Edington demonstrates an exceptional, almost robotic level of operational execution and capital discipline. By leveraging their core RF and mmWave competencies to secure and execute massive, binding contracts in the LEO space and defence sectors, management has consistently over-delivered on guidance without diluting shareholders, reflecting the highest tier of corporate integrity and competence. SK Hynix: 93/100 SK Hynix’s management team operates with absolute, ruthless operational efficiency, cementing a near-monopoly in the highest-margin sectors of the AI hardware supply chain. While their corporate governance reflects a traditional Korean chaebol structure—meaning individual executives hold virtually zero personal equity—their technical execution, capital discipline, and balance sheet mastery currently represent the gold standard in the global semiconductor space. $ASM.AS: 92/100 ASM International’s management team demonstrates exceptional execution, operational visibility, and capital discipline. While lacking the high insider ownership typical of aggressive, founder-led companies, their pristine balance sheet, brutally accurate multi-year financial forecasting, and active share buyback programs firmly establish them as highly trustworthy stewards of shareholder capital. $OUST: 90/100 Ouster, Inc. (OUST) presents a rare profile in the pre-profit / early-commercialization hardware space: a management team that actually executes on its roadmaps. Led by CEO Angus Pacala, the company has demonstrated strict operational discipline, successfully commercializing its digital lidar technology while maintaining a fortress balance sheet. While historical share dilution from strategic M&A requires monitoring, management’s transparency regarding one-time revenue events and their consistent delivery of hardware margins earn them an elite rating. $PL: 89/100 Planet Labs (PL) is a rare anomaly in the high-growth space: a former SPAC that actually survived its cash-burn phase and successfully transitioned into a free-cash-flow-generating enterprise. While management's equity compensation remains slightly elevated, their exceptional track record of launching highly complex hardware on schedule and their disciplined, jargon-free focus on securing massive government contracts demonstrates supreme execution and integrity. $MRVL: 88/100 Marvell’s management team operates with high technical competence and ruthless strategic clarity, successfully transforming the company from a broad-based silicon provider into a highly focused AI data center powerhouse. While management has faced cyclical headwinds in legacy segments that triggered some guidance volatility, their ability to aggressively hit engineering milestones for custom AI silicon and responsibly manage capital allocation reflects elite, tier-one execution. $PLTR: 87/100 Palantir Technologies (PLTR) features a visionary, high-conviction management team that consistently executes on commercial expansion and product delivery at a scale few enterprise software companies can match. While their aggressive stock-based compensation and slight equity dilution trigger mechanical deductions, their ability to drastically over-deliver on forward guidance and perfectly align capital deployment with their core Artificial Intelligence Platform (AIP) establishes them as elite operators. $TMDX: 87/100 TransMedics Group (TMDX) possesses an elite, execution-focused management team that has successfully transformed the company from a capital-intensive med-tech startup into a vertically integrated, cash-flowing logistics and clinical powerhouse. While minor deductions are warranted for recent share dilution and stock-based compensation levels, management’s flawless delivery on ambitious infrastructural timelines and mathematically transparent communication earn them near-maximum trust. $HIMS: 85/100 Hims & Hers Health, Inc. (HIMS) is operated by an aggressively competent management team that has successfully transitioned from a cash-burning SPAC into a highly profitable, cash-flowing entity. While historical share count expansion requires ongoing scrutiny, their ability to execute against stated timelines, beat forward guidance, and seamlessly expand their core direct-to-consumer moat into highly lucrative verticals borders on clinical precision. $RKLB: 79/100 Rocket Lab (RKLB) boasts a highly accountable, founder-led management team that excels in strategic focus and transparent communication, successfully transitioning the company into a space systems and defense prime contractor. While the persistent delays of their heavy-lift Neutron rocket require notable deductions, the company’s ability to consistently meet financial guidance and maintain technical integrity makes them a standout in the capital-intensive aerospace sector. $AMPX: 75/100 Amprius Technologies’ management team exhibits high-level execution regarding product iteration and revenue growth, but this comes at a severe cost to retail equity holders. While the team successfully transitioned the company toward profitability by ruthlessly pivoting to a fabless contract-manufacturing model, their aggressive share dilution and the quiet abandonment of their flagship domestic manufacturing plans reveal a management team that prioritizes survival and top-line growth over pristine shareholder alignment. $FLNC: 73/100 Fluence Energy (FLNC) presents a complex profile of elite financial backing and strong insider alignment, severely undercut by chronic project execution issues and lumpy guidance misses. While management operates with a fortress balance sheet and refuses to artificially pump the stock with trending AI buzzwords, their inability to consistently defend gross margins and deliver predictable quarterly revenues makes them a volatile, high-risk operator in the utility-scale energy storage sector. $IREN: 35/100 IREN (formerly Iris Energy) presents as a highly promotional, capital-intensive infrastructure play that relies heavily on retail shareholder dilution to fund its shifting ambitions. While management has successfully secured impressive debt financing and massive power capacities, their chronic inability to hit original timelines, massive expansion of the share count, and blatant narrative pivot from Bitcoin to Artificial Intelligence reflect a team that prioritizes growth at the direct expense of shareholder equity.

🚨BREAKING: $IREN ANNOUNCES SUCCESSFUL ENERGIZATION OF SWEETWATER 1 🔥🔥🔥🔥🔥🔥

@StormDirac @PerDahlstrm @Illskan It seems those who sold early — or never even joined the wagon — love to throw shade at those who remained invested. Human nature, I guess …

Kan tycka att dela information som enskild medborgare är inte pump, jag säger inte bra eller dåligt lyfter fram fakta, då det nu finns 100.000 tals intresserad utanför Sverige. Där emot att etablerad media helt utan att kolla på AI marknaden eller bolaget, fortsätter med rubrik efter rubrik (se bild från DI nedan). Det är något du borde ta upp som dump och inte ge dig på mig. Att sitta bakom ett anon konto och säga dessa saker är värt noll, våga stå för vem du är eller så kanske du evt vara del av blankarna som för söker dumpa, då är trovärdighet liten.

Win Semiconductors 실적 나왔네요. 제가 궁금했던 InP/포토닉스 쪽으로 생산이 확대되는 중! 이렇게 되면 $SIVE 진짜 매수해야 하나…? 매출 YoY 28% 성장, QoQ 4% 하락. 매출총이익률 26.3% (QoQ 5.5% 하락) 영업이익률 9.4% (QoQ 4.5% 하락) 숫자만 놓고 보면 그저 그렇지만 폭발적으로 성장할 포토닉스 / InP 레이저 및 광모듈 생산을 위해 캐파 증설 중이라는 컨펌! 의문/우려 가졌던 부분들이 해소되는 중입니다. 역시 $SIVE 질문할 시간에 풀매수 때렸어야 했나... 😅 @Sofigoodboy 소파이보이님 부럽...

@Frenchie_ excellent Discord, autant pour les novices que pour les expérimentés ! j'apprend beaucoup et la communauté partage beaucoup !

L’une de mes plus grandes fiertés aura été de créer STOCK FR une communauté centrée sur la finance, et plus particulièrement sur le stock-picking. Au sein de cette communauté, nous avons su identifier plusieurs tendances majeures : >Les semi-conducteurs depuis début 2025 >La montée en puissance des commodities, notamment l’or et l’argent >La photonique La diversité des profils réunis ici est aussi rare que précieuse L’objectif de cette communauté est simple : la recherche pure d’alpha Vous êtes les bienvenus

#SIVE $SIVE #5G #6G #Beamforming #FWA #SATCOM #CHIPSAct #Semiconductors #Wireless #Nokia Take a moment to read this, friends — we’re going to talk about the unknown side of Sivers. A lot of investors — and me first — only knew Sivers for its photonics business (the lasers for AI datacenters). It’s exciting and has massive potential. But after digging a bit deeper, we discover a real hidden gem: the Wireless division. It still accounts for ~70% of current revenue, delivered very strong growth (+85% on some quarters in 2025), and provides solid cash-flow stability. It’s the quiet but powerful pillar that makes the entire $SIVE investment thesis much less risky. First of all, what is a beamformer? (simple explanation) : Imagine you want to send a super-fast internet signal (5G, future 6G, or satellite connection) to a phone, a house, or a satellite. A regular radio signal spreads out in all directions, like a light bulb lighting up an entire room. It works, but it’s very inefficient. You waste a lot of energy, the signal weakens quickly over distance, and there are lots of interferences with other devices around. A beamformer (literally “beam former”) is like an ultra-precise smart projector: Instead of spreading everywhere, it concentrates all the signal power into a very narrow directional beam pointed exactly at your phone or your house. Result? The connection is much faster, much more stable, reaches much farther… and uses way less energy. Sivers Wireless develops exactly these mmWave beamformers (chips + antenna arrays). This technology is essential for high-speed 5G/6G, Fixed Wireless Access (FWA = wireless internet instead of fiber), SATCOM, and defense applications. Major wins to date (April 2026): • January 2025 → $5.4M contract (60 M SEK) signed with a global Tier-1 ( The crowd is murmering Nokia ? ) It’s for the next-generation beamforming transceivers. Even better: the collaboration is planned across multiple generations. • New → Firm $3M purchase order with Tachyon Networks (28 GHz FWA). Production and shipments already starting in 2026 — real revenue incoming. • 3 design wins in 5G/FWA already secured → potential of around 1 million units over the first three years of production (starting 2026). • India → Historic partnership with Intel + WiSig Networks + office opened in Bangalore to target the massive FWA market. • Defense → CHIPS Act funding (2nd round ongoing), contracts with Raytheon, BAE Systems, Ericsson. Daybreak™ 7-15 GHz ICs now available (March 2026) — very competitive in power and efficiency. • SATCOM → Cloudchaser chipset + Maverick antenna arrays launched (Feb 2026) + strong commercial interest. • 6G → Already involved in ISAC research with Nokia. Why is this strategic for $SIVE? Wireless brings stability and predictable revenues while the AI photonics division ramps into mass production (2026-2027). It’s the financial cushion that lets the company breathe… and it has strong potential to become a real cash cow from 2026-2028. And even if photonics hits some delays or doesn’t deliver as expected, the Wireless division alone is strong enough to justify a major re-rating on its own. In short: $SIVE is not just an AI laser story. With the recent news of the Photonics spin-off to Silicon Valley, $SIVE (Sweden) is effectively becoming a pure-play leader in Wireless/SATCOM/Defense, while still holding the majority of the US AI laser powerhouse. It’s the ultimate value unlock. It’s also a powerful play in mmWave 5G/6G, SATCOM, defense, and emerging markets — with technology that is already generating revenue. I hope this clarified some things and/or brought new info for the Sivers community! This is not financial advice. I’m just digging into this company and sharing what I found. #SIVE $SIVE #5G #6G #Beamforming #FWA #SATCOM #CHIPSAct #Semiconductors #Wireless #Nokia

#BloomEnergy #AI #Energy #DataCenters #Hydrogen #CleanEnergy #Investing 📷 Energy is no longer just a cost — it has become the #1 limiting factor of the entire AI revolution and global growth. In the years ahead, energy will be the real bottleneck… and the biggest opportunity. The explosion of artificial intelligence, the massive build-out of data centers, the electrification of transport and industry, and the urgent need to decarbonize are driving an unprecedented surge in electricity demand. Traditional power grids are already saturated, connection times are measured in years, and blackouts or volatile prices are becoming unacceptable. Without abundant, reliable, instantly deployable clean power, neither AI nor the energy transition can scale. Energy is the new oil. This is exactly where Bloom Energy is quietly becoming one of the biggest winners of the decade. 📷 Bloom delivers on-site, decentralized power with its industry-leading Solid Oxide Fuel Cells (SOFC) — the famous Bloom Energy Servers. These systems convert natural gas, biogas or hydrogen into clean electricity directly behind the meter, without combustion, providing 24/7 ultra-reliable power that can be deployed in just months instead of years. Perfect for hyperscale data centers, factories, hospitals and critical infrastructure that cannot afford downtime. What gives Bloom a massive competitive edge? → Efficiency >60% (up to 85% with heat recovery) — far superior to traditional turbines or other fuel cells → Fuel flexible (natural gas or biogas today, hydrogen tomorrow) → 99.999% uptime — “five nines” reliability → Modular, scalable and already proven at massive commercial scale Growth potential is explosive and still largely under-appreciated: • Multi-billion dollar backlog • Giant 2.8 GW contract with Oracle • Production capacity set to double • Major Brookfield investment (up to $5B) • Clear path to hydrogen economy + Power-as-a-Service recurring high-margin revenue Bloom doesn’t just produce energy — it makes the abundant, reliable, decarbonized energy future possible. On valuation: Even though the stock has already run hard, the current market cap still leaves a very big upside (“big X” still expected). It’s an extremely exciting long-term bet on the future of AI and clean energy. The street has not yet fully priced in the multi-year hypergrowth runway. Is Bloom the sleeper winner of the AI energy crisis? Drop your thoughts 📷

Merci pour l’analyse en live ! A mon avis, avec la redondance ils veulent montrer une volonté de proposer des solutions prêtes à déployer rapidement. Avec la soirée co-organisée NVIDIA + serveurs exposés sur deux stands (Submer + Valeo), c’est plutôt positif ? Peut-être un rapprochement plus structurel à moyen terme. A suivre…

2CRSi × Submer : ce qu'on apprend de l'OCP EMEA. Pendant que le marché digère un peu la hausse, parlons fondamentaux pour réduire notre émotionnalité : L'annonce principale, c'est un reference design conjoint avec Submer et ENEOS Group pour ce qu'ils appellent des « AI factories clé en main ». Concrètement : le hardware serveur est signé 2CRSi, le refroidissement par immersion liquide est signé Submer, et les fluides caloporteurs viennent d'ENEOS, le géant japonais de l'énergie. L'idée c'est de proposer une architecture complète, pensée dès la conception pour être déployée telle quelle. Mon interrogation c'est que que 2CRSi n'a pas attendu Submer pour faire de l'immersion cooling. Ils ont leur propre gamme, Atlantis, conçue en interne. Alors pourquoi aller chercher Submer pour construire un reference design sur exactement le même terrain ? Et ils ont déjà Chemours pour les fluides dielectrique donc encore une fois ENEOS est également redondant. Cette redondance-là me questionne. Historiquement, c'est exactement le type de configuration qui précède potentiellement un rapprochement plus structurel. Partenariat exclusif, prise de participation, intégration verticale je ne sais pas quelle forme ça prendrait ? Mais la logique industrielle est difficile à ignorer. Mais restons prudent là-dessus, c'est de la spéculation. L'autre signal à ne pas rater : la soirée technique de la veille a été co-organisée avec NVIDIA. Quand NVIDIA accepte de co-hoster un événement avec vous, c'est qu'ils considèrent que votre stack mérite d'être montré à leurs propres clients. Et il y'a ce détail : Sur place, les serveurs 2CRSi étaient exposés sur deux stands différents - celui de Submer - celui de Valeo. Deux partenaires, deux verticales distinctes. Submer pour le datacenter IA en immersion ? Valeo pour l'edge computing en environnements contraints ? Barcelone ne fait que confirmer ce qu'on lisait entre les lignes depuis des mois. Ce n'est pas un partenariat annoncé qui a été dans un communiqué de presse mais un produit fini, montré en live, devant les décideurs du secteur. Quand le cours corrige comme aujourd'hui et hier, c'est dans ces moments-là qu'il faut savoir où regarder. A suivre... 👀

Presenting 2CRSi In the current geopolitical context, Europe has no choice but to unite and build its own technological sovereignty to counter the hegemony of Trump’s America… and the serious candidates are few and far between. Let me introduce you to 2CRSi. 1/3 Who is 2CRSi? 2CRSi designs and assembles ultra-dense, high-performance servers (HPC / AI / data centers) with unique expertise in advanced liquid cooling (immersion, direct-to-chip, and two-phase).The company is also pivoting toward a Compute-as-a-Service (CaaS) model to generate more stable and higher-margin recurring revenues. Recent numbers: H1 2025/26 revenue ×9.8, raised guidance to >€400m revenue and >€36m EBITDA for the year, with an ambition of ~€1bn revenue in 2026/27.2/3 2/3 The ultra-bullish scenario If everything aligns: The AETHER Gigafactory project (40 MW, scalable to 300 MW), where 2CRSi is the technical coordinator, becomes a flagship European AI infrastructure project. Partnerships with Submer, Chemours, Valeo and contracts in the Middle East / India (NAFFCO) materialize. Strong margin expansion thanks to in-house servers + CaaS model. → Potential revenue of €3-5bn in 5 years, EBITDA margins 18-22%, market cap moving from ~€900m today to €4-8bn (x4 to x9). The future “European sovereign cooling Nvidia”?3/3 3/3 The bearish scenario Delays on Aether (permits, financing). Fierce American competition (Dell, HPE, Supermicro). Margins remain compressed if the mix stays too buy-resell GPU heavy. Small-cap volatility + dilution risk. Conclusion Despite the inherent risks of a small-cap, I remain very bullish on 2CRSi. The stock has already risen significantly in recent months, but the upside margin remains enormous. The combination of European AI sovereignty + liquid cooling leadership + CaaS pivot + concrete projects (Aether, Submer, etc.) is rare in Europe. On a 3-5 year horizon, the re-rating potential is one of the most asymmetric in the entire French tech sector.

Bullish for 2CRSi !! Maybe something big incoming ?