Tech Observer

$2,000,000 in a month is something I never thought in my wildest dreams would be possible. I continue to be shocked not just by the market we are in, but how predictable it is and the duration of the run. Yes I am fully aware this will not last forever, the investing style I have taken as of late is not long term sustainable, but I am adapting to the moment and trying to take full advantage. Previously I had an account with a lot of followers I would post investment thesis, research philosophies and even AI investment training materials…. I am actually very proud of those as I know I helped a lot of people. But the account became a barrier to me using some of my strongest market intuition skills, being agile, operating quickly on new information and being in tune with the market. Nothing pisses off followers more than someone who posts an investment thesis then changes their mind shortly after. Just look at our friend @daniel_koss, who is called a grifter by many because he regularly pivots, but somehow manages to continue booking meetings with executives of public companies… both can’t be true. And I’m pretty sure the public company executives can figure out which is which. I found myself holding onto positions longer than I wanted because I felt I owed it to my followers to stay even when I would have rather moved to names with strength that were in trend at the moment. In a market like this, moves happen fast, and chasing strength has been consistently rewarded. X seems to over-value accounts sticking with investments and resent those who move quickly… even if the latter significantly outperforms. I’m pretty sure my bank will take my checks from my quick moves. Here is what I see in the market moving forward. 2025 and early 2026 rewarded the small/mid cap emerging companies who became critical parts of the ai cycle. Companies like $be $crdo $alab $nbis $iren $aaoi $mu and others came out of nowhere to become some of the most important in the world. Those of us who were early made explosive gains taking risk while others called it a bubble. We layered into options and had 500%-1000% years. I believe the rest of 2026 is about leaning into strength with leverage. Memory, connectivity and energy will continue to outperform. Winners will grow by 50-100% and we already know who they are. There are now simple tools to add leverage. 2x ETFS can hurt you but they are simple and offer high liquidity and strong returns. If you’re in tune with the market you can use them to outperform. I don’t see the value in trying to find the next small cap to explode right now, that was 2025. Missing the moves of the clear winners just isn’t worth chasing those high risk opportunities. Personally my head is in one place…. Can I get one more double. $10,000,000 means I’m done forever. Most of this is in my Roth so it’s tax free, I’m still 39 so can’t access most of it but I found a tax code (72T) that allows me to pull before 59 without penalty if I want…. So I basically have a call option on it. Overall I don’t think we will see as many explosive moves as we saw the last year, but using leverage on the cleaner opps offers a lower risk way to get the same value. Frankly I think it’s easier. Last piece. I’ve got a tiny following here. I just want to be clear that I plan to move quickly and aggressively with little explanation for it. If you follow a move of mine please understand I could fully unload that position 24 hours later. I don’t want any misconceptions about what I’m doing right now. I’ve been researching this space essentially 24/7 for 18 months and I’m just trusting my conviction weekly on where the best value for my cash at any given moment is. I’m really just rotating about 15 names as charts and themes get hot or cool off. I see no reason to stop. I really hope everyone is taking advantage of this. It’s a life changing opportunity that we won’t see again for a long time.

$MU I feel like people are underweighting that Micron is the only one of the big three memory players that is American. A lot of things would have to change very quickly for this not to be materially higher by EOY. The floor here is very high.

@_HalalTrader_ InshaAllah!

$MU what a beautiful a dragonfly doji! We’re going much higher InshaAllah!

Hedge funds sold the most tech stocks in a decade last week. Could be viewed as a canary in a coal mine for further downside… I actually think this is a local bottom signal. We will know more by end of July.

@PodcastAlphaX This page is incredible. You’re going to have a billion followers. Incredible work.

In last 7 days we went through 11 pods. Here's what stood out across them: - Enterprise AI ROI: who's getting returns and who's running pilots - AI CapEx: two credible people with opposite 2027 calls - Foundation model economics: token cross-subsidy, commoditization, talent - Crypto + tokenization: independent convergence on RWA - AI in biology: a Nobel laureate's verdict on AI washing - Macro: consumer behavior vs. sentiment, Iran pipeline, rates - Investing framework: structural traps that haven't gone away Very excited to share these findings with you all as Podcast Alpha has grown to 7500+ followers in a very short time. 🙏

“I believe AI capex will rise 40%-60% next year” - @munster_gene For context the street is projecting 22%-30%. If true, this means companies like $MU (and other core infra names) have a lot more runway that is not priced in. This sounds insane on its face, but when you look at a few data points from the weekend this starts to make sense. 1. @ExponentialView released an incredible study showing that Hyperscalers are beginning to see enough ROI to cover the depreciation on their AI infra spend. Doesn’t mean cash flow positive but it means this capex “bubble” is becoming sustainable with a path to improved outcomes. Can’t overstate what this means. 2. @FinancialTimes published a report stating @Google had to turn down massive compute requests from Meta over the last year due to capacity constraints. What this means: The ai bubble talk is generally based on the idea that the spend cannot continue to grow at this pace. However if the hyperscalers are truly showing the math is backing out… we could be getting looking at a sustained increasing spend for 5-10 more years. 🔹The math in the @ExponentialView report says this is on the horizon. 🔹 Google turning down customers like Mets says the demand is deeper than the street believed. 🔹 Micron locking down 5 year deals with the majority of their customers tells us the buyers believe this to be the case. I fundamentally believe we are still in the first 2-3 innings of the build out. Not because I am an AI bull… Because that’s what the data says.

@azeem Thank you for linking this. Look forward to reviewing in detail.

@J5jayfive intelligence.exponentialview.co

This is the most important story of the weekend, ignore the memory noise. “Revenue from artificial intelligence has reached a tipping point, showing that the hundreds of billions of dollars tech companies are spending on it may be economically sustainable”

@CStats9 Yes. Holding all the 2x is not a long term strategy. Need to be aware of the volatility decay. I only use it when I think a stock is at a dirt cheap valuation or showing technical signs of a breakout.

@J5jayfive Are you not worried about the 2x leveraged … if MU status at this price and consolidates for 2 or 3 weeks , wouldn’t that effect more in terms for MUU

$2,000,000 in a month is something I never thought in my wildest dreams would be possible. I continue to be shocked not just by the market we are in, but how predictable it is and the duration of the run. Yes I am fully aware this will not last forever, the investing style I have taken as of late is not long term sustainable, but I am adapting to the moment and trying to take full advantage. Previously I had an account with a lot of followers I would post investment thesis, research philosophies and even AI investment training materials…. I am actually very proud of those as I know I helped a lot of people. But the account became a barrier to me using some of my strongest market intuition skills, being agile, operating quickly on new information and being in tune with the market. Nothing pisses off followers more than someone who posts an investment thesis then changes their mind shortly after. Just look at our friend @daniel_koss, who is called a grifter by many because he regularly pivots, but somehow manages to continue booking meetings with executives of public companies… both can’t be true. And I’m pretty sure the public company executives can figure out which is which. I found myself holding onto positions longer than I wanted because I felt I owed it to my followers to stay even when I would have rather moved to names with strength that were in trend at the moment. In a market like this, moves happen fast, and chasing strength has been consistently rewarded. X seems to over-value accounts sticking with investments and resent those who move quickly… even if the latter significantly outperforms. I’m pretty sure my bank will take my checks from my quick moves. Here is what I see in the market moving forward. 2025 and early 2026 rewarded the small/mid cap emerging companies who became critical parts of the ai cycle. Companies like $be $crdo $alab $nbis $iren $aaoi $mu and others came out of nowhere to become some of the most important in the world. Those of us who were early made explosive gains taking risk while others called it a bubble. We layered into options and had 500%-1000% years. I believe the rest of 2026 is about leaning into strength with leverage. Memory, connectivity and energy will continue to outperform. Winners will grow by 50-100% and we already know who they are. There are now simple tools to add leverage. 2x ETFS can hurt you but they are simple and offer high liquidity and strong returns. If you’re in tune with the market you can use them to outperform. I don’t see the value in trying to find the next small cap to explode right now, that was 2025. Missing the moves of the clear winners just isn’t worth chasing those high risk opportunities. Personally my head is in one place…. Can I get one more double. $10,000,000 means I’m done forever. Most of this is in my Roth so it’s tax free, I’m still 39 so can’t access most of it but I found a tax code (72T) that allows me to pull before 59 without penalty if I want…. So I basically have a call option on it. Overall I don’t think we will see as many explosive moves as we saw the last year, but using leverage on the cleaner opps offers a lower risk way to get the same value. Frankly I think it’s easier. Last piece. I’ve got a tiny following here. I just want to be clear that I plan to move quickly and aggressively with little explanation for it. If you follow a move of mine please understand I could fully unload that position 24 hours later. I don’t want any misconceptions about what I’m doing right now. I’ve been researching this space essentially 24/7 for 18 months and I’m just trusting my conviction weekly on where the best value for my cash at any given moment is. I’m really just rotating about 15 names as charts and themes get hot or cool off. I see no reason to stop. I really hope everyone is taking advantage of this. It’s a life changing opportunity that we won’t see again for a long time.

@Romolooo @SemiAnalysis_ Research firm. Poorly stated. Fair critique.

@J5jayfive @SemiAnalysis_ Since when semianalysis is an investmend fund? Do you even understand what you write?

This is Dylan Patel. He runs @SemiAnalysis_, an investment fund focused on AI Infra. This is from early April, it’s when I went all in on Micron $MU. You can listen to 30 year fund managers on CNBC, I listen to this guy. If you can’t tell the difference, can’t help you.

@Chaimeisenberg lol. Research / consulting firm. Forgive the misnomer. Im a fan. Keep up the good work.

All this time I’ve been working at an investment fund and had no idea

@nickvannewkirk Why are you being positive on X. Have you been in this platform? You’re supposed to call me an idiot. Haha. Thanks man.

@J5jayfive You’re killing it. Been awesome reading back thru your old posts. Congratulations on the trade so far!

Everyone is focused on $MU and $AAPL This is 10X more important for the AI trade If Hyperscaler debt for capex has positive ROI… which this says it is and will improve. We are looking at an infinite money glitch for the next 5 years where the massive spend becomes extended.

@Jueyegof I would argue the signal speaks more to the depth of the supply crunch. Maybe it’s not nothing… but if the first alternative is a chinese company literally on a banned list… that’s not an awesome signal for the market and where else it can be bought

@J5jayfive HBM represents a quarter of their revenue or 35% of DRAM. Having one of your largest customers looking to replace you and open up supply is not good for your product that only benefits on a supply constraint. Doesn’t mean the world’s ending for micron but it’s not nothing.

Finally someone who posts the actual reality of the situation. 1. Speaks to desperation and how long this supply shock will be. 2. Does not impact the business $MU makes real money from. 3. Highly unlikely.

@NuguardGroup You know more about memory and HBM manufacturing than Gavin Baker?

@J5jayfive HBM is not rocket science. CXMT could enter that business if they wanted to and they can afford the cost to develop those production lines.

I don’t know who needs to hear this but $CXMT does not and cannot make the type of high margin memory $MU supplies to hyper scalers. Even if they get approved to supply $AAPL, they are just taking some low margin business that frankly $MU would love to do less of.

@rubicon59 This is the real story of the weekend. If this Bloomberg story is true and they are seeing positive ROI on debt, we might have an infinite money glitch for a while… The memory bears will at some point have to submit to math and physics.

Still true today IMHO $DRAM

@Awhartd @eugene_dol35177 Possible…. But currently (as of today) Bloomberg is reporting that Capex spend is seeing positive ROI. I would consider updating your mental model. This is not the same memory market as previous cycles. Look at the real data my friend.

@J5jayfive @eugene_dol35177 The next nail in the coffin is when when one big cap tech cuts capex or pushes it out due to high memory prices. It’s coming. Repost the end of the day MU is a commodity company and will follow the boom bust cycles of them.

@CHAITANYA5 @SemiAnalysis_ They can’t make this product. Literally. They can supply Apple. Meta/goog/amzn

@J5jayfive @SemiAnalysis_ Cmxt is coming.

@sam90860759 No one else can make it dude. There is no where else to go. It’s a physics/math issue. Not politics.

@J5jayfive Ya but if one window is open what’s stopping google, amzn etc to find different supplier?

Google is paying out the 🍑 for compute from XAI. Cloud leaders are currently turning customers away becuase they can’t meet demand. This is a land grab for compute. And now we see positive ROI. The spend has a long way to go to close the gap. This is the math I see when running the models on ai hyperscaler capex… 80% still in front of us. And demand is proving this is needed

@RehanYawar84469 @J5jayfive @SemiAnalysis_ Yes. Look at the deals that xAI has been signing.

Which dot com bubble companies had revenue like this and went to the bottom? You’re conflating companies that did nothing with one that will be sitting on hundreds of billions in cash. Also…what intervention are you suggesting is possible. There are 3 companies that can make this product. It’s virtually impossible to enter the space. This is a physics and math issue.

@J5jayfive @eugene_dol35177 Heard the same type of logic from guys that road the dotcom bubble all the way down. This is very negative for the DRAM companies. If you think this stops at Apple you are crazy. This is just the opening salvo from the much more politically connected tech companies against MU.

@OpenClawCentral @RehanYawar84469 @SemiAnalysis_ FWIW… this is more important than any memory news. If roi on hyperscaler capex is positive…. We are looking at a medium term infinite money glitch during this buildout.

@J5jayfive @RehanYawar84469 @SemiAnalysis_ Sanjay micron ceo: (paraphrase) Robotics will create a multi Decade demand cycle for memory. No one is bullish enough. $MU $DRAM