かい

$MP - 34 days until Section 122 tariffs are due to expire (July 24) under the 150-day clock that started Feb 24. It's gotten messier since: a trade court ruled the 10% tariff unlawful on May 7, but the Federal Circuit stayed that ruling on June 11 and said the government is "likely to succeed" on appeal, so the tariff is still being collected today. Congress hasn't enacted an extension either way. If the surcharge lapses, imports still face normal MFN rates plus whatever Section 301/232 tariffs already apply, not a clean reversion to zero. What it means for $MP holders: this legal fight is mostly noise next to the real catalyst, the DoD equity partnership from 2025. Tariff headlines might jolt the stock short-term, but the structural story (US trying to onshore rare earth supply with the Pentagon as a direct stakeholder) doesn't change whether this surcharge survives appeal or not.

$NBIS (Nebius) Nebius is exiting 2026 with up to 1,000 megawatts of data center capacity online. They started the year at 170 megawatts. That's close to a 6x jump in twelve months, backed by a contracted backlog approaching $50 billion through 2031. What it means for NBIS holders: the stock isn't pricing a neocloud story anymore, it's pricing whether they can physically build power and chips fast enough to fill contracts already signed. What is your EOY Price target?

1/ Quick $NBIS thread on the one number that actually decides who wins with these AI cloud names, and barely anyone tracks it. Not revenue. Not GPU count. Power. Heres why locked up gigawatts matter more than chips, and what Nebius just quietly grabbed 🧵 2/ Heres the thing that matters. GPUs arent really the bottleneck anymore. You can order H200s. What you cant just order is power and land at scale. A gigawatt takes YEARS to permit and build. So whoever locks up power now owns the capacity in 2027. Everyone else is stuck in line waiting. 3/ And this is where Nebius has been quietly winning. Theyve already contracted over 3.5 GW of power, going for 4 GW by year end. And back in May they grabbed 1.2 GW of power plus land for their own AI factory in Pennsylvania. Owned, not rented. Thats years of runway locked in. 4/ Then the part almost nobody noticed. Just last week, June 10, they closed the acquisition of Eigen AI, an inference and model optimization company. Why care? Cause a pure GPU landlord gets commoditized fast. Moving up into inference is how they protect margins instead of racing everyone to the bottom on price. 5/ And the demand backs all of it up. On the last call they straight up said theyve raised prices AGAIN and are still selling out across every chip type. Thats not a company struggling to fill racks. Thats one that cant build them fast enough. 6/ Im not blind to the risks though. This is a capex monster, they guided $20 to 25B and keep raising convertible notes, so dilution and financing risk is real. A lot of revenue leans on Meta and Microsoft, thats concentration. And the stock is up triple digits this year and swings like crazy. But that power moat is the part the market still sleeps on 🔌

1/ This Fed meeting was a way bigger deal than the flat headline makes it look. They held at 3.5 to 3.75%, but the real story is buried in the projections. Let me break the numbers down 🧮 2/ Start with the dot plot. The median rate call for end of 2026 jumped from 3.4% in March to 3.8% now. Thats a big move. The committee basically went from pricing a cut to pricing a 25bp HIKE by year end. 9 of 18 members now see at least one hike. Only 1 still sees a cut. 3/ And the why is all in the inflation data. CPI ran 4.2% in May, hottest since 2023. The Fed yanked its end of year PCE forecast up from 2.7% to 3.6%. GDP got trimmed to 2.2%. So slower growth AND hotter inflation. Thats the uncomfortable combo for them. 4/ Heres the technical bit almost nobody mentions. Nominal rates look high at ~3.6%. But measured against 4.2% CPI, the real rate is actually slightly negative. So policy isnt nearly as restrictive as the headline number feels. Thats part of why risk assets havent fully rolled over here. 5/ Now the valuation mechanics. The discount rate on future cash flow just went UP, not down like the market was betting on. Higher discount rate means lower present value means multiple compression. And it hits long duration growth names the hardest, since most of their value sits way out in the future. 6/ The front end is the quiet one. Short T bills are paying around 4% risk free right now. That sets the hurdle rate for everything else. Any risk asset has to clear a guaranteed ~4% just to justify the risk. That math has been sitting under spec names all year. 7/ So step back. Cuts pushed to 2027. A hike back on the table. Real rates near zero. Risk free at 4%. None of that shows up in a single earnings report, but it shapes every multiple in your portfolio. Watch the macro as close as the tickers. Hike or hold next print? 👀

Most people scroll right past the Fed headline and have no idea how much it actually moves their stocks. Quick rundown. The Fed held rates at 3.5 to 3.75% on Wednesday. Theyve been steady all year, last cut was back in December. It was Kevin Warshs first meeting as the new chair and the tone shifted a bit hawkish. They pulled the easing language out and the dot plot now leans toward a possible hike this year, with any cuts pushed out to 2027. Why? Inflation picked back up. The Iran war sent oil higher in Feb and CPI hit 4.2% in May, the highest since 2023. So the Fed is in no rush to cut. What it actually means for stocks, simply: Rates are the gravity behind every valuation. Higher for longer tends to matter most for long duration growth names priced on future cash flow, and less for companies already making real money today. And T bills paying around 4% give investors a safe place to sit, which is always part of the math. Not bearish, not bullish, just the setup. A big chunk of what moves your portfolio comes from up here, not from the company itself. Worth watching as close as you watch your tickers. Hike or hold from here? 👀 $AMD $NBIUS $ASTS

1/ Everyone is losing their minds over $MSFT right now. Stock got crushed, lawsuit headlines everywhere, people screaming the AI spend is reckless. Meanwhile Bill Ackman quietly threw 2 BILLION into it at like 21x earnings. Called it a highly compelling valuation. Crowd panics, that guy buys. Im paying attention 💵 2/ So why is it even this cheap. Wasnt the earnings, those were fine. It was the capex. They guided ~$190B for 2026, up like 61%, way over what the street wanted to see. Everyone freaked about free cash flow and the multiple got chopped. Nothing about the actual business broke tho. 3/ And heres what the bears keep forgetting. Nobody leaves Microsoft. Ever. Once your whole company runs on Office, Teams, Azure and Active Directory, youre stuck. 450M+ users locked in. You dont rip all that out cause some startup has a shiny new tool. The switching cost is insane. 4/ And the AI demand is as we know real, its sitting in the numbers. Azure did 40% last quarter, sped back up, beat their own guide. AI run rate over $37B, up 123%. Only thing slowing Azure down is they literally cant build data centers fast enough. 5/ Now the real question. Does $190B a year in AI spend actually pay off, and how fast? Right now demand says yes. But its a huge bet. If the returns come slower than people want, this thing can chop sideways and be boring for a while So yes there is a risk. 6/ And Im not ignoring the rest either. ~45% of that $627B backlog leans on OpenAI. They burn cash and can now spend on AWS and Google too. Real concentration there. Copilot adoption still slow, like 4-5% paying. Smart money is split too. Ackman loaded up, TCI dumped most of theirs scared of AI eating Office. Not a clean one way bet. 7/ Where Im at. Most embedded software and cloud company on the planet, trading at a multiple it basically never sees, real AI demand under it, and Ackman buying right next to me. The capex risk is real. The OpenAI thing is real. But Im holding every share here. What are you guys doing with $MSFT?

$MSFT just got sued by shareholders after a $357B market cap wipeout. The accusation is simple: investors claim Microsoft made Azure look stronger than it really was, while not fully showing how much AI capex was starting to pressure the business. The lawsuit says Microsoft allegedly hid slowing Azure growth, overstated the strength of its AI story, and downplayed how expensive the AI buildout was getting. After earnings, the stock dropped around 10% in one day, so shareholders are arguing they bought the stock at an inflated price.

1/ $MSFT is the cheapest its been in over a decade and honestly Im kinda loving it here. 22x earnings. The last time I saw it this cheap was the 2022 bottom. And the thing nobody around me is talking about? This company owns 27% of OpenAI. Let me tell you why Im holding every share into the OpenAI IPO 🧵 2/ So why is it this cheap. I think the market is scared of two things. One, the whole "AI kills software" panic, and software is most of what MSFT does. Two, theyre throwing $190B at capex this year and its eating the free cash flow alive. Thats how the multiple went from 31x down to 22x. I dont think either one actually breaks the long game. 3/ Heres what I keep coming back to. After the restructuring last year MSFT owns ~27% of OpenAI. At OpenAIs last valuation thats worth around $228B to them, off an investment of about $13B. I dont think most people pricing this stock are giving them anywhere close to full credit for that. 4/ And this is the part I really want you to pay attention to. OpenAI filed for its IPO in June and theyre gunning for over $1 trillion. The day that lists, MSFTs cut gets a real public price tag, like $270B+. Thats basically 10% of the whole company sitting in one number the market keeps treating like its zero. Im not treating it like zero. 5/ Its not just the OpenAI bet either. Azure ripped ~40% last quarter and the only thing slowing it down is how fast they can physically build data centers. They got a $627B backlog. They locked OpenAI into $250B of Azure spend. The core business is still a beast, I dont know how else to put it. 6/ Am I ignoring the risk? Nah, I own this thing so I gotta be honest with myself. That same OpenAI stake means MSFT eats part of OpenAIs losses too, and theyre burning like $27B a year, so it drags earnings right now. The capex is real. And a chunk of that backlog leans on OpenAI who isnt even profitable yet. Im not calling it free money. 7/ And Im not gonna lie to you on timing. OpenAI literally said the IPO "may be a while," and the betting markets have it around 40% to even happen this year. Could easily slip to 2027. So Im playing this as a when not if, not a next week thing. Im not trading it, Im holding it. 8/ But real talk, I get to own the most locked in software and cloud company on the planet at a 2022 multiple, that also owns a quarter of the most hyped startup ever, right as that stake is about to get repriced live. I added on this dip. I dont get setups this clean on a mega cap often. What are you guys doing with $MSFT here?

@Novariv_ It's a whazy. It's a woozie. It's fairy dust.

1/ $ASTS dropped almost 16% the same week they literally put 3 more satellites in orbit. Think about how dumb that is for a second. The company is hitting the biggest milestones in its history and the stock is getting smoked. I keep seeing people asking what happened and if the thesis is dead, so let me actually break this whole thing down. Spoiler, Im not selling a single share 🛰️ 2/ The drop has almost nothing to do with the company. On June 12 SpaceX IPO'd at $1.77T, biggest listing ever, and basically every dollar that was parked in small speculative space names got yanked out and thrown at the new shiny thing. The whole sector bled, not just us. $ASTS just got hit the hardest, and theres a reason for that. SpaceX literally named $ASTS as a competitor in their IPO filing. So when the big dog goes public and points at you, you become the first thing people sell. Add in the index funds being forced to buy SpaceX and you get a giant capital vacuum. None of that is $ASTS doing anything wrong. 3/ Then on top of the rotation, we did the classic sell the news. They launched BlueBirds 8, 9 and 10 on June 17, clean success, and the stock faded anyway cause everyone bought the rumor and dumped the event. And look, Im not gonna sit here and pretend theres zero real overhang. Q1 revenue came in soft, insiders have been selling, and $SPCX just did a $17B spectrum deal with EchoStar that makes their direct to cell push stronger. That EchoStar one actually matters and I'm watching it. But none of this is the business falling apart. Its sentiment, rotation, and one competitor headline. 4/ Now heres what everyone forgets the second the candle turns red. $ASTS is the ONLY company building a cellular network in space that talks straight to a normal phone. No special case, no dish, no extra hardware. The phone already in your pocket. Each BlueBird Block 2 is a 2,400 sq ft monster antenna, one of the biggest commercial arrays ever flown, and it can push up to 120 Mbps down to a regular handset. Thats the moat. Nobody else is doing this at scale. 5/ And before anyone calls it a meme stock burning to zero, look at the actual balance sheet. $3.5B in cash. Over $1.2B already contracted. 2026 revenue guided at $150M to $200M. 50 carrier agreements covering close to 3 billion subscribers, were talking AT&T, Verizon, Vodafone, T-Mobile. Plus 8 development contracts with the US DoD. This is a funded company with the biggest carriers on earth already signed, not some hope and prayer ticker. 6/ The part that actually gets me excited is the model flipping on. Right now its mostly a build story. But as they get sats up and turn on commercial service, the carrier revenue share kicks in, and you go from burning cash on a constellation to actually printing recurring revenue off 3 billion potential subs. Theyre at 9 operational sats now and targeting around 45 by year end. That cadence is the entire game. 7/ Am I ignoring the risks? Absolutely not, thats how you blow up an account. Starlink direct to cell plus that EchoStar spectrum is a legit competitive threat. They HAVE to hit the ~45 satellite cadence by year end or the whole multiple comes down. Theyre still deeply unprofitable and burning $1.6B+ in free cash this year. And the insider selling isnt a great look. If they miss on launches, I'll be the first one to say the thesis cracked. But missing hasnt happened yet. 8/ So heres where I land. I never thought this thing rips straight up, its a high beta build and you have to be able to stomach 40% drawdowns like the one were in right now. But the core didnt change this week. First mover in direct to phone satellite broadband, fully funded through the build, carriers already locked, sats going up on schedule. The price got caught in the SpaceX flush. The story didnt. Down 40% from the high and Im still holding every share. What are you guys doing here? 🛰️

Back in October 2025 they already had capacity to produce 6 satellites per month. This is likely a confirmed DOD or USGov or EUGov contract and/or AST building another LEO shell in parallel. Eventually ASTS will have 3+ LEO "shells" for different functions, frequencies, customers.

$ASTS under $100 isn't the company breaking. SpaceX IPO'd at $1.77T and pulled every speculative space dollar into one name. ASTS got hit hardest, it's the direct Starlink rival. Meanwhile: 3 new BlueBirds up, $3.5B cash, 45 sats by year end. I'm not selling this dip 🛰️

$KRKNF regulatory approval is in. Covelya closes July 2. Let me lay out the actual math, because the market still hasn't. THE DEAL Kraken announced Covelya on March 3 for C$615M (C$480M cash, C$135M stock). The cash was funded by a bought deal that priced at C$8.50, upsized to C$402.5M, plus a C$150M credit facility. That C$8.50 is where institutions committed real money. The stock trades below that today. (all figures CAD) WHAT KRAKEN BECOMES Before this, Kraken was the power and sonar inside undersea drones. Now it owns the navigation, comms, positioning and optical layers too, through Covelya's six subs: Sonardyne, EIVA, Forcys, Wavefront, Voyis, Chelsea. The full internal stack, regardless of who builds the hull. THE NUMBERS Kraken standalone 2025: ~C$103M revenue, ~C$25M adj EBITDA, cash flow positive, scaled from C$25.6M in 2021. Covelya 2025: ~C$249-275M revenue, ~24% EBITDA margin, growing 24% a year since 2023. Combined 2025: C$365M revenue at a 24% adj EBITDA margin. 2026 Kraken organic guide: up to ~C$175M, roughly 60%+ growth, before Covelya even hits the books. This is a profitable, growing business. Not a story stock burning cash. THE ANDURIL ANGLE Sonardyne was already supplying Anduril's seabed sentry line before this deal. So now every Ghost Shark, every Dive-LD, every REMUS unit isn't just batteries and sonar from Kraken. It's nav, comms and optics too. Same programs, more Kraken content per vehicle. VALUATION Kraken is profitable at ~24% combined margins and growing 60%+ organically, yet still trades like a tiny venture name. Half the buzzy drone and defense tech names trade at far richer multiples on little or no profit. You're paying a discount for the better business. CATALYSTS Covelya closes July 2 and management drops combined guidance. A TSX main board or NASDAQ uplisting unlocks the funds that structurally can't touch a venture listing today. Institutions are only ~23-26% of the register, retail holds ~70%+. Uplisting flips that gate. A named Anduril program award with confirmed scope turns the whole narrative into hard numbers. THE RISKS (because they're real) 47M+ new shares means dilution. Integrating a C$615M deal is not trivial. Covelya has heavy customer concentration, one customer was ~45% of 2025 sales. Margins could revert and the Anduril content could come in smaller than the bulls hope. Institutions paid C$8.50. It's cheaper now, it's profitable, and it's growing. I know which side I'm on. 🦑 As always, make your own due diligence, research!

$HOVR Most eVTOLs waste energy with exposed rotors. Horizon Aircraft’s patented HOVR Wing is different. The fans are hidden inside the wings. Once it transitions, it flies like a normal airplane for ~98% of the mission. This gives it better efficiency, speed, and range. They already have strong partnerships: ✅ Motion Applied (ex-McLaren F1) ✅ Pratt & Whitney Canada ✅ MT-Propeller ✅ Marshall Aerospace Next major catalyst: Full-scale Cavorite X7 prototype assembly targeted for late 2026, with first flight testing in early 2027. Still early, but the setup looks promising. Some are targeting $10–20 if they execute. Watch how the tech works here: youtube.com/watch?v=hzYZdx…

$KRKNF $PNG.V regulatory approval is in. Covelya closes July 2. This is the deal that turns Kraken from a parts supplier into a full subsea system integrator. Power, nav, comms, sonar all under one roof. $615M CAD, combined rev ~$365M. Who else holding? 🦑

Told y'a :)

For new investors trying to learn stocks properly 👇 Here are 5 things that actually helped me (instead of just gambling): 1. Reddit + X are goldmines… but dangerous ones. Join the specific subreddit of the stock you’re researching. You’ll find real discussions. But always remember, a lot of people on social media are just trying to pump their bags or stay bullish for no reason. Take everything with a grain of salt. 2. Learn how to actually read a company. Don’t just watch YouTube videos. Learn how to read a balance sheet. Understand what EBITDA, Forward P/E, and CAPEX actually mean. This is the most important skill. Without it, you’re just guessing. 3. Use AI as your research assistant. I regularly use Grok, Claude Opus, and GPT in agent mode to dig through filings, news, and risks. They’re surprisingly good at summarizing complex stuff quickly. However always make sure to check the sources. Yes.. I know it takes time but that's what it takes to invest in Individual Stocks. 4. Keep it simple when you’re starting. As a new investor, don’t overdiversify. Start with 1-3 non-overlapping ETFs max. Only start buying individual stocks once you have strong conviction and are mentally prepared to lose money on them. 5. The hardest (and most important) skill: Learn to not panic sell. Build real conviction by researching the management team and even checking Glassdoor reviews. The employees are usually the best indicator of how a company is actually run. Also… sometimes just touch grass for a few weeks. Your portfolio will survive. What’s one thing you wish you knew earlier when you started investing?

6/7 However ‼️A career fair does not put satellites in orbit.
AST has missed timelines before, and they already trimmed the 2026 target from 45 to 60 satellites down to around 45.

7/7 And the stock already prices in a lot. So even good news might not move it much. (like yesterday bluebird launch). 
But if you believe the thesis, a 40% headcount jump to support a 6 satellite per month line is exactly the kind of signal you want.
Not financial advice.