Shay Boloor
@StockSavvyShay • 458,302 subscribers
Chief Market Strategist @FuturumEquities | Regular on @Reuters, @YahooFinance, @Bloomberg, @FoxBusiness, @SchwabNetwork & @Forbes | NIA
Shorts
Videos
Mark Zuckerberg says “the value of even an H100 at this point is higher than it was last year” as AI compute remains scarce challenging the idea that older $NVDA GPUs quickly lose their value. Thats also why Meta benefits from owning its data centers because it can keep using GPUs long after they're fully depreciated instead of paying cloud providers year after year. The same logic extends to $NBIS, $CRWV and $IREN because if GPUs remain productive longer than expected then they can generate more revenue from the same hardware without needing to replace it as quickly.
Shay Boloor93,314 görüntüleme • 6 gün önce
Everyone is chasing stocks that have already run while $MELI is still trading near its 2021 highs despite revenue growing to a ~$40B annual run rate. What also makes Latin America so interesting is consumers already have smartphones and spending power yet e-commerce in Brazil is still only ~12% of retail while financial services remain underserved. Thats why I think $MELI and $NU have years of growth ahead without needing to expand beyond Latin America because both are building the commerce and financial infrastructure for demand that already exists.
Shay Boloor76,774 görüntüleme • 5 gün önce
Cathie Wood says she avoids memory stocks like $MU and $SKHY because she still sees HBM as the most cyclical and commoditized part of the semiconductor stack. Her bigger point is that inference architectures from $CBRS and $NVDA via Groq can reduce or eliminate the need for HBM which makes today’s memory pricing surge look more temporary than structural. As she puts it when technology costs “triple or quadruple” engineers eventually design around the bottleneck and she believes inference is already starting to do exactly that.
Shay Boloor734,610 görüntüleme • 2 ay önce
I just made $CRDO a top 5 position in my growth portfolio because Credo is evolving into a much broader bet on owning the connection inside AI clusters. As those clusters move toward 1.6T and eventually 3.2T then Credo can capture more of the link across electrical, optical, DSPs and silicon photonics as bandwidth and distance requirements increase. And once you own more of the link then products like Pilot let Credo move up another layer by monitoring connection health before failures start dragging down utilization across increasingly expensive AI clusters.
Shay Boloor512,028 görüntüleme • 1 ay önce
$PLTR Rule of 40 score reached 155 and the metric has now expanded for 12 consecutive quarters as operating leverage continues improving with scale. That means Palantir is growing close to 100% while converting more than half of revenue into operating income which is a combination almost unheard of in large-scale software. $GOOGL, $AMZN, $MSFT and $META are spending hundreds of billions building the AI factories while Palantir sits above that infrastructure and turns the resulting intelligence into real-world outcomes without funding the physical buildout allowing far more incremental revenue to flow into profit and cash.
Shay Boloor688,460 görüntüleme • 2 ay önce
The $NBIS, $CRWV and $IREN debate comes down to three very different competitive advantages: • Nebius is most interesting higher up the stack where software around training, inference and agents could increase value of every megawatt it deploys • IREN wins on power access, site development and ability to bring large amounts of AI capacity online efficiently • CoreWeave wins on operating experience with proven ability to run massive AI clusters for some of the most demanding customers in the market. The real question is which advantage proves hardest to replicate as competition increases.
Shay Boloor292,795 görüntüleme • 1 ay önce
The next 12 to 18 months could be when scale of the $IREN opportunity finally shows up in the numbers. That opportunity is already taking shape as IREN turns scarce power into ~$4B of contracted ARR tied to 2026 capacity with 3-year contract pricing up about 125% and active discussions reaching $25M per IT megawatt.
Shay Boloor334,836 görüntüleme • 1 ay önce
Zenzap just launched AI agents that work inside employee group chats to qualify leads handle approvals and schedule meetings across business apps. So essentially instead of employees switching between apps to get work done, agents can now handle those tasks directly inside conversations while people oversee the process.
Shay Boloor37,055 görüntüleme • 4 gün önce
I still own $IREN because I think the market is underestimating both the $NVDA alignment and how hard its power footprint is to replicate. If the company can turn ~5 GW of secured power into Nvidia-aligned AI cloud capacity then the market may start valuing the 10x gap between its power base and near-term AI cloud needs.
Shay Boloor1,314,709 görüntüleme • 5 ay önce
UPDATED GROWTH PORTFOLIO EXPOSURE Performance: • Growth Portfolio: +431% (93% CAGR) • $QQQ: +171% • $ARKK: +154% • $SPY: +91% Exposure: • AI Hardware | 22% • Space Economy | 20% • Digital Platforms | 13% • Healthcare | 9% • Physical AI | 9% • AI Power | 7% • Drones | 7% • Quantum | 6% • Power Grid | 1%
Shay Boloor1,374,141 görüntüleme • 5 ay önce
$AVGO already controls ~75% of AI ASIC market with six core customers including $GOOGL, OpenAI and Anthropic which Hock Tan says becomes its largest custom chip client in FY27 as agentic inference drives more workloads toward custom silicon. The part I like most is shipments are expected to double to ~10M in 2027 and reach 24M by 2030 while ASPs rise from ~$10K to ~$18K as chip complexity increases. So if AI revenue really moves above $100B in FY27 then ~20x earnings looks ridiculous for a business still scaling custom compute and the networking around every cluster.
Shay Boloor116,959 görüntüleme • 15 gün önce
Jensen Huang brushed off OpenAI’s claim that its $AVGO built Jalapeño chip beats GB300 by saying “I don’t have to take anything personally.” $NVDA is still gaining AI market share which is why Jensen says he remains “very comfortable with all the competition.”
Shay Boloor362,669 görüntüleme • 1 ay önce
The Fed keeping rates higher through 2027 is a real headwind for $CRWV because they already have to build so much capacity well before the revenue shows up where net interest expense alone hit $640M last quarter. But $NBIS raising GPU prices ~20% across both older and newer generations while still adding capacity tells us the same AI compute market CoreWeave is selling into remains very tight with demand still absorbing supply faster than expected. So I do think the Fed made CoreWeave’s financing burden worse today while Nebius gave us another strong signal that demand needed to support all that investment remains very strong.
Shay Boloor213,346 görüntüleme • 1 ay önce
$MU gave us a lot more evidence that memory could get tighter before it gets better with management saying 2027 and 2028 may be more constrained than 2026 even after the huge move in pricing. Whats wild is that Micron still expects revenue to rise as price increases moderate which means growth can shift toward more volume and better mix instead of needing DRAM up high teens and NAND up 30% every quarter. That’s why the story is getting bigger than HBM because HBM actually hasn’t been Micron’s highest margin product yet so 2027 repricing can improve that while AI demand keeps pulling on HBM, server DRAM and enterprise SSDs across entire memory stack.
Shay Boloor105,106 görüntüleme • 17 gün önce