@babyfolio Yeah. Happy with it, since I’m trying to build my short term account up to be 20% of my total portfolio value. Halfway there. So the rest of $SIVE will run now. Getting an entry into $PENG or $FCEL with good results could probably bring me to 15-20% if they perform and SIVE and ASTS keep going aswell. Trying to hit a 80/20 portfolio balance in terms of long-term // short-term.
@babyfolio 1% of total port. Just sold 50% of my SIVE position with 200% profit — after this mornings 20% up. The rest will run. So will probably add $PENG to my short-term account OR $FCEL if entry is OK.
A bit of FUD on $PENG today from @aleabitoreddit. The FUD-campaign from @aleabitoreddit on $IREN haven’t been a success so far. Food for thought. -BP Not financial advice.
I'm 100% sure if I met all you "photonic memory" experts in real life. 498 out of 500 of you couldn't explain CXL memory pooling or KV cache infrastructure and what $PENG actually does to derive revenue off that. This is why I'm seeing all these random $RKLB, $HIMS, or non technical AI experts on my timeline now. Backseat commenting completely wrong things about M7U MOCVD capex and $TSEM that aren't related. Or conflating every single term like an $SMCI integrator with photonics IP. Then just pitching buzzwords every under every one of my posts.
Very nuanced. People buying $PENG for $MRVL "photonic memory" are likely to be disappointed. They're on the $SMCI integrator level with potential software add. The high margin, foundational IP belongs to companies like Celestial or others like Lightmatter. What they do is build a 2U box, inside the rack, around Celestial/Marvell's the photonic memory IP. It's also in the development collaboration stage. Retail just misunderstand the different layers and conflate them all as this company building core photonic memory IP. Not commenting on the potential price, but it's a legitimate idea for its base business. And upside would be material if they go past sampling. Just annoying when I get comments about "PHOTONIC MEMORY". Then they're building the chassis that the actual photonic memory IP sits inside and not even in qualification stage.
5/6. Three wildcard picks that fits his frameworks for the next AI bottleneck: Wildcard 1: $CRDO Build the cables and chips that move data between GPUs inside AI clusters. revenue up 272% year over year. 88% market share in active electrical cables. the plumbing of every large-scale AI cluster running today. market still treats it as a niche semi name. Risk: copper faces long-term pressure from optical. customer concentration is real. Wildcard 2: $ALAB Builds the silicon that connects GPUs, CPUs, and memory so they can talk to each other at scale. Q1 2026 revenue up 93%. CXL memory controllers shipping to hyperscalers now. As agentic AI runs continuous multi-step workflows, memory-sharing becomes structural. astera built the controller for it first. Risk: valuation leaves no margin for error. any hyperscaler moving in-house re-rates it hard. Wildcard 3: $PENG 40 years of memory expertise. now building CXL-based servers that store AI agent context so inference doesn’t restart from scratch every step. Integrated memory revenue up 53% YoY. AI business now 60% of sales. A tier 1 bank is already running their product. Risk: Advanced computing revenue dropped 42%. Thin margins. Smaller operator in a space.
$308 million in integrated memory revenue. First half of fiscal 2026. Same window, a year ago: $201 million. $PENG didn’t just grow a segment. It rotated the ENTIRE business. The market has been pricing this as a hardware integrator in managed decline and the top-line supports that read. Revenue down 6% year-on-year last quarter. Advanced Computing collapsed 42%. Hyperscale hardware, exited. That framing is not wrong. It’s just looking at the half of the business that’s shrinking. The other half is a DIFFERENT picture. > Integrated memory: $308 million in H1 FY2026, up 53% year-on-year. > Per the Q2 FY2026 8-K filed April 1: that segment now represents 45% of total company revenue. The market has been pricing the decline. The data from this week describes something else. In March, at $NVDA GTC, Penguin launched the MemoryAI KV Cache server. Per the Q2 earnings transcript: the first production-ready CXL-based KV cache server commercially available. 11 TB of memory per node. 10x faster than NVMe-based caching. A Tier 1 financial institution has already deployed it. CXL competitors; SK Hynix, Astera Labs have demonstrated components. Penguin shipped a turnkey system. That is not the same category. Tuesday this week, Penguin disclosed AI-driven revenue now exceeds 60% of total first-half sales. Up from roughly 40% a year prior. Full-year guidance raised to 12% growth at the midpoint. Non-GAAP EPS to $2.15. Per the 8-K: $489 million in cash and short-term investments. No major debt maturities until 2029. Most people are still looking at $PENG and seeing a declining systems integrator navigating a messy transition. But when you look at 53% integrated memory growth, a first-mover CXL inference product with a live Tier 1 deployment, and AI revenue crossing 60% of the business… It’s interesting how those three things point at the same thing. The data is starting to describe a different kind of company. When more of the inference deployment cycle comes into view, I’ll post it here. -BP Not financial advice. Do your own due diligence.
$ONDS is up 31% today. Here’s my next bet trading at a discount right now: Most people on X missed $ONDS today. Not because they didn’t know about it. Because they SOLD it. I watched people unload 50% of their position the night before earnings. Fear of a bad print. Fear of volatility. Fear of being wrong. And I get it. Protecting capital is wise. If you genuinely fear the outcome, reducing risk is the right call. But here’s what I also saw: The investors who held through horrible $RKLB $NBIS $ONDS $IREN $CIFR chop without flinching? They all had one thing in common. They knew exactly what they owned. They didn’t need the price to tell them if they were right. They already knew from the filing. From the contract list. From the thesis they built months ago. That’s the edge nobody talks about. Not the scanner. Not the chart pattern. The ability to stay calm when everyone else is making emotional decisions. That calmness is not personality. It is PREPARATION. The best investors I’ve met on here have sat through the hardest chops of the past year without panic selling. Not because they’re fearless. Because they did the work before the red day came. Which brings me to my next bet. I’m long $PENG. GPU race dominates headlines. Post-purchase deployment does not. $PENG owns the last mile: full-stack AI infrastructure for enterprises lacking internal teams. > OriginAI: pre-validated GPU clusters + ICE ClusterWare orchestration, built on 3.3B+ GPU runtime hours. > 30+ years memory expertise via CXL MemoryAI KV cache server. > Up to 11TB expandable memory for inference, breaking GPU memory wall. > End-to-end: hardware, software, memory, deployment, management/turnkey production scaling. > 8–9 analysts: Strong Buy/Buy > 76% > FY26 revenue guidance raised to 12% growth. > Trades 16x forward PE (sector 23x). > Hyperscaler build-out priced in. Deployment layer underneath is not. -BP Not financial advice. Do your own research.
🚨THE BEARS HAVE BEEN WRONG ABOUT THE SAME THING FOR THREE YEARS. And they’re about to be wrong again. Morgan Stanley just raised their S&P 500 target to 8,300. Year-end 2026 target: 8,000. Twelve-month target: 8,300. Over 12% upside from current levels around 7,400. But here’s the line that matters most: “Our bullish index view is an earnings story, not a multiple expansion one.” Read that again. EARNINGS. > 83.2% of S&P 500 companies beat Q1 estimates. > Morgan Stanley projects $339 EPS for 2026. That’s 23% growth, followed by $380 in 2027 and $429 in 2028. A compounding earnings machine. Not a bubble. Now let me show you what the bears keep getting wrong. There’s a clip making the rounds right now. A guy pulls up two charts side by side. > Left chart: The dot-com era. Price ripping higher. Earnings flatlined. Pure narrative. Pure multiple expansion. Pure speculation. > Right chart: Right now, 2023-2026. Earnings are LEADING price. AI and semiconductors are driving actual profits. Price is still CATCHING UP to the gray line. His point? We’re not in a bubble. We’re in the middle of an earnings cycle that hasn’t fully repriced yet. He’s right. $MU, $SNDK, SK Hynix and Samsung, 75% of the $DRAM market are proof. These are companies printing money faster than analysts can revise targets upward. > The memory supercycle isn’t priced in. > The AI infrastructure buildout isn’t priced in. The earnings are real and the multiples are still compressing INTO the growth. The dot-com bubble was speculation in search of earnings. This is earnings in search of a price that can catch up. There’s a massive difference. So where does the money go? $SPY and $QQQ as a passive bet + into the infrastructure that’s CAUSING the earnings surge. Here’s where I’m positioned: → $IREN → $NBIS → $CIFR → $AAOI → $SIVE → $RKLB → $OUST → $PENG → $ONDS → $KRKNF Every single one of these names sits inside the structural AI buildout that’s driving the future earnings Morgan Stanley just upgraded their entire index outlook for. HERE’S THE MACRO SETUP Morgan Stanley explicitly ties their revised outlook to AI adoption enhancing operating leverage across the S&P 500 and a rolling earnings recovery that continues to progress. This isn’t a macro call. This is a capital allocation call. The earnings cycle is real. The infrastructure buildout is real. The defense spending acceleration is real. $QQQ and $SPY tell you the tide is coming in. The names above tell you which boats rise the most. The dot-com bubble was price without earnings. This is earnings without price. The gray line hasn’t been touched yet. We’re mid-cycle. Act accordingly. Still long. Still adding. Still building. -BP Please remember: This is not financial advice. Do your own research. I hold positions in many of the names mentioned.
$NVDA x US x China agreement. My thesis is; this is probably in exchange for Iran peace. If I expect market to fly, and my strategy will be to take some profit, before a dip. Lets go $IREN $NBIS $AAOI $OUST $PENG $CIFR $RKLB -BP Not financial advice.
🤯WOW! I actually think this might be my best day across both my portfolios. Long-term: 10% $NBIS $AAOI $OUST $CIFR $RKLB $IREN $PNG.V $ONDS Short-term: 20% $PENG $SIVE Here’s the screenshot. You know what that means 😂 Sorry. -BP Not financial advice, https://t.co/vmXlC7QAFR
PORTFOLIO UPDATE If you followed me this year, your short-term portfolio is up 125%. Your long-term book is up 55%. I don’t use options, calls, leap, short e.g. The $SPX has spent most of 2026 trying to figure out what it is. We have not. WHY I SHOW UP EVERY DAY 1st January I had 1,207 followers. Today we are approaching 23,000. No subscription. No paid community. No X payout. No product. Just showing up every day and doing the work. I made a pact with my daughter to show up every single day. To show her what consistency actually builds. Not in theory. In real time, with a number she can watch move. I have a good job. I don’t need to monetize this. What I need is to show her what happens when you commit to something and don’t quit. By EOY, I want her to see 50K on that number. I’m still far from it. I also know how FAST that can change when the content connects and the market moves. So I keep going. @Sandeman52 is someone I think about here. He built something real without noise. That’s the benchmark. I’m here for the stocks, the connection, and the game. Everything else is secondary. LONG-TERM PORTFOLIO Eight positions. All structural. Sold $AMPX today with 30% profit. Added to my positions in $KRKNF, $OUST and $ONDS $RKLB: +83% | 18.51% of port. $NBIS +142% | 17.09% of port. $IREN +55% | 13.61% of port. $AAOI +113% | 12.11% of port. $OUST +39% | 11.60% of port. $CIFR +42% | 10.41% of port. $KRKNF -17% | 9.14% of port. $ONDS -1% | 7.52% of port. SHORT-TERM PORTFOLIO: $PENG +10% | 66.1% of port. $SIVE +98% | 33.9% of port. HOW I SEE THE REST OF 2026 Three forces are running the tape: 1. The AI tsunami is still in the first inning. Only 1% of companies consider themselves mature AI users. Over 92% plan to increase AI investment. McKinsey estimates cumulative US data center spending alone will reach $5 trillion by 2030. That capital is already committed. The infrastructure phase is being priced. The productivity phase hasn’t arrived yet. The application phase is after that. I don’t see a dot-com bubble. The dot-com companies burned cash on speculation. The companies in this cycle have real revenue, real backlog, and real physical constraints that large capital cannot route around. 2. Political trade has replaced free trade. This is the structural shift most retail investors are still underweighting. Real technology, real projects, real contracts. Political decision in Washington and repriced overnight. That is the new operating environment. Capital allocation now has a new first-order variable: political geography. Which government wants this to succeed? The US-UAE AI Acceleration Partnership, $1.4 trillion committed, sovereign wealth choosing the US as the headquarters of the next industrial era, is the clearest current signal. Political trade creates volatility in names exposed to the wrong jurisdiction. It creates structural advantage for names embedded in the right one. 3. The economy is shifting from consumption-driven to production-driven. The post-WWII consumer economy ran for 80 years on demand expansion. What is building now is a production expansion cycle, driven by onshoring, defense spending, AI infrastructure, and energy security. The companies that build, enable, or supply that production cycle are not being valued for what they are today. They are being valued for what the production economy needs them to be in 2028 and beyond. That is what this portfolio is built on. MY POSITIONING FROM HERE I expect more volatile than consensus expects. Full of buying opportunities for anything with hard assets and contracted revenue. Adding on dips. Not selling on headlines. The three forces above are not quarterly variables. They are decade-long structural shifts. Volatility between now and December is the mechanism that creates the next entry points or DCA opportunities. YTD +55% long-term. +125% short-term. And the cycle has barely started. -BP Note: This is not financial advice. I hold positions in all tickers mentioned.
15 of my favorite AI infrastructure plays in this market right now: $MU $AMD $INTC $SNDK $AAOI $MRAM $PENG $MXL $CRDO $MRVL $GLW $NOK $CEVA $GFS From memory, networking, photonics, optical connectivity, foundry, edge AI to telecom infrastructure — this sector is becoming the backbone of the next AI cycle. AI demand is no longer just about GPUs, it’s power, memory, data movement, fiber, cooling, networking and low latency infrastructure. That’s where I see some of the biggest long term opportunities developing. Every pullback in strong AI infrastructure names continues to get bought aggressively so far. Selective accumulation still looks like the best strategy in this market IMO.
PM looks insane! 🤯 Its only a couple of days since I posted my thesis on $PENG and took a position. $PENG is smashing +20% in PM 🟢 And Wall Street is about to wake up. Credit for $PENG goes to @pennycheck. Happy to share the ride with other first movers; @FinnStockinger @michaelsikand @CKCapitalxx Others: $AAOI above $200 now. 7% up in PM 🟢 $SIVE showing strength with 13% 🟢 $RKLB taking off with 7% 🟢 $NBIS hammering 4% 🟢 $OUST taking on 5% 🟢 I’m working on a extensive portfolio update. Tag along. Follow. Stay tuned. Onwards, -BP Reminder: This is not financial advice.
@zizonchain @aleabitoreddit @ParadisLabs Love it man! Also shared it 👊 Btw, have you seen $PENG OVERNIGHt?
$PENG $MRAM $MXL $CEVA $NOK Some of the best AI infrastructure plays in this market right now. AI infrastructure is not just GPUs anymore. Memory, edge AI, connectivity, photonics, inference, low power compute and telecom infrastructure are becoming the next wave of the AI buildout. $PENG quietly becoming an AI factory / enterprise AI infrastructure story with HPC, memory and cluster management exposure. $MRAM is one of the most underrated AI memory themes. MRAM technology is gaining traction for low power AI compute, edge AI and in-memory computing applications. (Tom's Hardware) $MXL positioning for AI networking and connectivity demand. $CEVA strong edge AI and wireless IP exposure as AI moves toward devices and edge computing. $NOK very interesting long term AI infrastructure + telecom + data center connectivity play as sovereign AI and AI networking spending grows globally. Market still heavily focused only on hyperscalers and GPUs, but the next leg of AI money flow could move deeper into infrastructure, memory, networking and edge AI names. Risk/reward in these types of early infrastructure plays can become asymmetric very fast.
$PENG is quietly becoming one of the more interesting AI factory plays in the mid-cap space. Penguin Solutions designs, builds, and manages end-to-end AI infrastructure — think GPU-scale deployments, HPC platforms, and inference-focused solutions under its OriginAI brand. The stock is +20% on the week, fueled by a raised FY2026 outlook projecting 12% net sales growth on the back of surging enterprise and government AI factory demand. Rosenblatt just bumped their PT to $54 (from $32) and kept their Buy rating after management meetings. With ~$489M cash, price-to-FCF around 9x, if the AI infrastructure buildout continues accelerating, $PENG could be one of the overlooked names in the picks-and-shovels stack. Not financial advice.
$PENG up +18% in AH. I hoped you listened and followed me. -BP Not financial advice. https://t.co/GuyQyt0WOj
$PENG https://t.co/yOO8MyYCWG
$MRAM and $PENG AH moving
Let me remind you of this today: There’s no bubble. Stop whining and being a 🐱. We are going higher when retail 🐱 is out. $IREN is a $200 stock. $NBIS is a $400 stock. $CIFR is a $100 stock. $AAOI is a $600 stock. $RKLB is a $500 stock. $OUST is a $100 stock. $PENG is a $150 stock. Stop worrying. Breath. Relax. Imagine where we are in 2030. Long term mindset. Don’t let them shake you out. That’s how strong markets, stay strong. -BP Reminder: This is not financial advice.
Higher, higher, HIGHER! 😎 Let’s go bulls! Sadle up $IREN $NBIS $AAOI $PENG $OUST $RKLB. -BP Not financial advice.
My thesis on $PENG: _ Beautiful short-film created by @zizonchain Starting: @aleabitoreddit @ParadisLabs @pennycheck
Wall Street haven’t figured $PENG role out yet.. when they do, it will explode. Here’s everything you need to know and why $NVDA $AAOI $MU $SNDK will need to rely on them, at one point: $PENG provides the *end-to-end architecture* required to make components work as a unified system. They design, build, and manage "AI Factories" such as massive data centers optimized for AI training. Their portfolio covers: > Computing > Memory > Managed Services They solve the "Memory Wall" (the bottleneck between the processor and data) by owning memory technology and system integration expertise. And acts as the integrator layer, the "glue" that assembles raw components into a functioning supercomputer. Penguin’s uniqueness lies at the intersection of Memory and HPC (High-Performance Computing). Most integrators focus only on the compute (GPUs), but Penguin understands that the "Memory Wall" (the speed gap between the processor and the data storage) is the BIGGEST bottleneck in AI. They are one of the few players that own the memory technology (through their SMART Modular brand) and the system-level integration expertise. To give you an understanding of their position: Tier 1: Component Providers (The Raw Materials) $NVDA provides the Blackwell GPUs that serve as the engines of the factory. $MU provides High-Bandwidth Memory (HBM), while $MRAM provides persistent memory for reliability. $SNDK provides high-speed SSDs, and $P provides the all-flash arrays required to feed massive datasets to GPUs without lag. $AAOI provides the "nervous system" through 800G and 1.6T optical transceivers that move data between GPUs. $POET and $SIVE provide photonics to replace copper wires with light-speed links. $LPK provides precision manufacturing for the circuit boards these components inhabit. Tier 2: The Architect & Integrator (The Brain) This is the center. Penguin takes the chips $NVDA, memory $MU, storage $P, and networking $AAOI and "weaves" them together. They perform the "racking, stacking, and testing". Tier 3: Infrastructure Operators $NBIS & $IREN. These are the customers. They build the physical power, shells, and the AI cloud. They rely on Penguin for the technical deployment and managed services to keep fleets running 24/7. If the AI revolution is a gold mine, $NVDA and $MU are the pickaxes, $P and $AAOI are the conveyors and bins, $IREN and $NBIS are the ‘owners’, and $PENG Solutions is the engineering firm that builds and manages the entire operation. -BP NOT FINANCIAL ADVICE.
@aleabitoreddit Can’t honestly believe you haven’t mentioned $PENG yet. https://t.co/akHwth5HTU
I took a position in $PENG today. Here is why and my PT: > 2026: $93 base / $124 bull > 2027: $160 base / $210 bull From $44 today. That is a 5x if the re-rate completes. The mechanism is not earnings growth. It is multiple expansion EV/Revenue moving from 1.6x toward 4-5x as the market stops calling $PENG a hardware company and starts calling it what it actually is. The same re-rate drove $MU +800%. The same re-rate drove $SNDK +4,000%. $PENG is a $2.5 billion company that owns the only production-ready solution to the single biggest performance bottleneck in enterprise AI. The market is pricing it like a server reseller. Here’s the actual chain reaction, and where the money is: Layer 1: The GPU layer (what everyone sees) $NVDA B300s. $AMD MI300Xs. Already priced, already covered, already bought. Layer 2: The DRAM layer (what sophisticated investors see) $MU and $SNDK supply the HBM and NAND underneath every inference workload. Real bottleneck, real trade. Also fully committed, hyperscaler supply agreements have absorbed production capacity through 2027. Not much left for the market to price in. Layer 3: The inference memory architecture layer (what nobody sees) AI inference is 70% memory-bound. Not compute. Not storage. Memory. A GPU running enterprise inference idles constantly; the on-chip HBM fills up, the key-value cache has nowhere to go, the GPU stops and rebuilds. That idle time is the MEMORY wall. It is not theoretical. It is happening inside every enterprise AI cluster being deployed right now. $MU supplies raw DRAM. $SNDK supplies NAND. Neither ships a system that fixes the architectural problem inside the cluster. Nobody does. One company excepted. > $PENG > Market cap: approximately $2.5 billion. > The MemoryAI appliance delivers 11 TB of CXL memory per node. Speeds 10x faster than NVMe caching. Native NVIDIA Dynamo compatibility and plugs directly into any NVIDIA inference cluster without touching the software stack. > Patent-pending. > Already deployed at a Tier 1 financial institution at production scale as of Q2 FY2026. The supply chain looks like this: SK hynix DDR5 → CXL Add-in Cards → Penguin MemoryAI Server → NVIDIA Dynamo KV offload → GPU inference cluster Penguin owns Layer 3. Astera Labs sells CXL controllers, components only. XConn and MemVerge have demonstrated lab architectures but have not shipped a turnkey server. Penguin ships the COMPLETE appliance, backed by 3.3 billion GPU runtime hours that no competitor entering this space in 2026 replicates in under 18 months. The financials are not what the price implies. Q2 FY2026: $343M revenue. Non-GAAP operating margin 13.2%. Gross margin 31.2%. Expanding year-over-year even as total revenue fell, because Integrated Memory grew 63% YoY. Net cash positive: $489M cash, $450M debt with no maturities until 2029. Buyback: $32M deployed in Q2 alone at approximately $19 per share. Management was buying at less than half today’s price. $112M authorization still remaining. FY2026 guidance raised May 10: 12% net sales growth. $2.15 Non-GAAP EPS. Now the valuation. $PENG trades at approximately 1.6x EV/Revenue. $MU traded at 1.8x before the inference re-rate. Moved to 5x. 800% return. $SNDK at 2x as a storage company. Reframed as AI inference infrastructure. Moved to 8-10x. 4,000% return. $PENG is at 1.6x. Priced like nobody noticed it shipped the only production-ready inference memory system in the world. 2026 bull: $124, 4x EV/Revenue. Category recognition. 2027 base: $160 — 4x on ~$2.0B as Integrated Memory sustains 35%+ YoY. 2027 bull: $210 — 5x on ~$2.1B. Full platform recognition. The catalyst: Q3 FY2026 earnings, approximately July 2026. $375M guided. $0.55 EPS guided. If MemoryAI deployments are named, the re-rate begins. -BP Not financial advice. @ParadisLabs @KawzInvests @pennycheck
$MU $AMD $INTC $P $MRAM $MXL $CEVA $PENG $HIMX $ON $GFS $GLW $NOK Some of the better-looking setups in this market right now. Semiconductor, connectivity, memory, optical, and AI infrastructure names continue showing relative strength while many sectors are chopping sideways. A lot of these charts are either breaking out or setting up near key levels with momentum still favoring the sector rotation theme.
I’m considering building a new big LT position within AI infrastructure constraints. Power, memory, substrates and now glass… There’s SO many opportunities circulating right now; $LPK $P $PENG $DGXX $MRAM just a few examples. The industry is moving SO fast. So, what’s the best cases out there in your perspective? I need inspirations for my research.
$PENG Penguin Solutions: Navigating the AI Infrastructure, Memory Pivot, and Investment Thesis. Penguin Solutions is in the early stages of repositioning from a legacy hardware provider toward AI infrastructure and memory platform solutions. The strategic direction is clear, but the current financials don't yet reflect the pivot — the recent earnings beat was driven primarily by cyclical memory pricing rather than the Advanced Computing segment, which actually declined. That's an important distinction: the headline looked good, but the business mix the market is underwriting hasn't shown up in the numbers yet. The core debate is straightforward. Is this a genuine AI infrastructure story in progress, or a hardware and LED business with a compelling narrative overlay? The answer matters significantly for valuation, and the evidence so far tilts more toward the latter than the former. Leadership has been refreshed. The new CEO brings enterprise experience, which is the right profile if the target is inference workloads and AI factory solutions — customers in that space buy differently than traditional hardware buyers, and enterprise sales motion matters. Whether that translates into execution improvement is still to be determined. The stock has moved. Rapid price appreciation suggests the market has already assigned meaningful probability to the AI pivot succeeding, which compresses the upside and raises the cost of any execution misstep. This is now a prove-it setup — the valuation requires the Advanced Computing segment to not just stabilize but grow, and for that growth to carry the margin profile consistent with an AI infrastructure platform rather than a commodity hardware vendor. The path to justifying the current price runs directly through margin conversion. Technical expertise is present, but durable margins in this segment require repeatable, high-value solution sales — and that capability has not yet been demonstrated at scale.
$PENG all-time high. https://t.co/WCSiaGzyau
$PENG — Penguin Solutions $PENG is an AI data center infrastructure specialist that beat Q2 expectations on both revenue and operating income despite a 6% year-over-year revenue decline, with the real story being a markedly more bullish full-year guidance raise from new CEO Kash Shaikh — driven by strong memory demand and five new AI/HPC customer wins. The company operates across Advanced Computing, Integrated Memory, and Optimized LED segments, offering HPC and AI solutions under the Penguin Computing and Stratus brands — serving OEMs, enterprise, and government customers through direct and third-party channels. Analysts project a path to $1.8B in revenue and $316M in earnings by 2028, requiring ~10% annual revenue growth — ambitious but credible if the AI infrastructure buildout continues. Goldman Sachs remains a Buy while Barclays recently downgraded to Equal Weight with a $27 target — a split Street reflecting the tension between a compelling AI tailwind and near-term execution. Not financial advice.
