🇨🇭 Global One SI - Prompt
$HIVE/USD H2 #Indicator #Signal #Trading #GlobalOneAI
HIVE Digital's $HIVE BUZZ HPC secured a $220M, three year GPU cloud deal with Bell Canada and Cohere.
The partnership deploys 2,304 Nvidia $NVDA Grace Blackwell GPUs in British Columbia to power sovereign AI workloads for enterprise clients.
The partnership deploys 2,304 Nvidia $NVDA Grace Blackwell GPUs in British Columbia to power sovereign AI workloads for enterprise clients.
🇨🇭 Global One SI - Prompt
This media is not supported in your browser
VIEW IN TELEGRAM
Traders trying to predict the market
Trump & Iran
Trump & Iran
🇨🇭 Global One SI - Prompt
Traders trying to predict the market Trump & Iran
It’s a lot of mind games now I’m getting migraines 😂
This media is not supported in your browser
VIEW IN TELEGRAM
FIFA forcing players to have multiple hydration breaks to get a few ads in.
🇨🇭 Global One SI - Prompt
$TRT/USD H2 #Indicator #Signal #Trading #GlobalOneAI
$TRT is pretty cheap compared to $AEHR.
People also talk a lot about its supposed link to $COHR.
And to other giants like $AMD, $NVDA, and $MU.
So I went deep and wrote my DD.
I kept it simple.
No jargon, no filler, straight to the point.
What it is:
A 67-year-old company that stress-tests chips before they ship.
Picks and shovels.
It gets paid no matter which chip wins.
And revenue is exploding: +58%, then +82%, then +124% the last three quarters.
The AEHR gap, honestly:
TRT trades at a fraction of AEHR.
That gap is massive, and it IS the opportunity.
Because if it weren't, it would already be priced higher.
You'd already be late.
It wouldn't be an opportunity anymore.
Yes, the two don't do the exact same job (AEHR is wafer-level at 40-50% margins, TRT is board-level at ~16%).
That's the bear case, and it's fair.
But that's the whole point: the ONLY thing between TRT and a re-rating is that margin.
And TRT looks like it's running thin margins on purpose, taking high-volume work to get in the door, with high-margin boards and a brand-new plant lined up to flip it.
The day that margin turns, the gap doesn't slowly close.
It snaps shut.
The connections:
One is already locked in: NXP, on its official supplier list. TRT is inside a tier-1 chip supply chain, on paper.
And the trail doesn't stop there.
$AMD, $NVDA, $MU and $COHR all turn up around TRT: job ads, shared landlords, sites next door.
Nothing officially confirmed yet, but the smoke is everywhere.
Yes, these are rumors.
But not astrology.
Tangible, real things.
Coincidences that might be coincidences, or might not.
And the fact they're still rumors is exactly what makes this a buying opportunity.
If it were already official, you'd be late.
Buy the rumor, sell the news.
The AI customer, though?
That one is real. Just unnamed, for now.
The whole thesis in one number:
Gross margin fell from ~26% to ~16%, because the new growth is low-margin AI testing.
For years it held 23-27%.
If that margin recovers, you get earnings growth AND a re-rating at once.
That's the 10x case.
If it doesn't, the price already reflects a lot.
Penang:
They just signed a big new plant as chips leave China.
Right now it's basically empty.
A fixed cost, today.
The bet is it fills with high-margin work.
Or probably already has a customer's name on it?
I wouldn't be surprised.
We'll see.
The risks (real ones):
Thin margins. One customer is ~41% of sales.
No clear moat yet.
The plant is a cost before it's a profit.
Why I'm long anyway:
In the middle of an AI boom, with capacity maxed out and chips forced out of China, I don't buy that TRT is cutting margins for no reason.
I think it's getting in cheap to land giants and earn the margin later.
High risk, high reward.
The test that settles it: next earnings in September.
Watch if revenue AND margin rise together.
Long $TRT. Not financial advice. DYOR.
People also talk a lot about its supposed link to $COHR.
And to other giants like $AMD, $NVDA, and $MU.
So I went deep and wrote my DD.
I kept it simple.
No jargon, no filler, straight to the point.
What it is:
A 67-year-old company that stress-tests chips before they ship.
Picks and shovels.
It gets paid no matter which chip wins.
And revenue is exploding: +58%, then +82%, then +124% the last three quarters.
The AEHR gap, honestly:
TRT trades at a fraction of AEHR.
That gap is massive, and it IS the opportunity.
Because if it weren't, it would already be priced higher.
You'd already be late.
It wouldn't be an opportunity anymore.
Yes, the two don't do the exact same job (AEHR is wafer-level at 40-50% margins, TRT is board-level at ~16%).
That's the bear case, and it's fair.
But that's the whole point: the ONLY thing between TRT and a re-rating is that margin.
And TRT looks like it's running thin margins on purpose, taking high-volume work to get in the door, with high-margin boards and a brand-new plant lined up to flip it.
The day that margin turns, the gap doesn't slowly close.
It snaps shut.
The connections:
One is already locked in: NXP, on its official supplier list. TRT is inside a tier-1 chip supply chain, on paper.
And the trail doesn't stop there.
$AMD, $NVDA, $MU and $COHR all turn up around TRT: job ads, shared landlords, sites next door.
Nothing officially confirmed yet, but the smoke is everywhere.
Yes, these are rumors.
But not astrology.
Tangible, real things.
Coincidences that might be coincidences, or might not.
And the fact they're still rumors is exactly what makes this a buying opportunity.
If it were already official, you'd be late.
Buy the rumor, sell the news.
The AI customer, though?
That one is real. Just unnamed, for now.
The whole thesis in one number:
Gross margin fell from ~26% to ~16%, because the new growth is low-margin AI testing.
For years it held 23-27%.
If that margin recovers, you get earnings growth AND a re-rating at once.
That's the 10x case.
If it doesn't, the price already reflects a lot.
Penang:
They just signed a big new plant as chips leave China.
Right now it's basically empty.
A fixed cost, today.
The bet is it fills with high-margin work.
Or probably already has a customer's name on it?
I wouldn't be surprised.
We'll see.
The risks (real ones):
Thin margins. One customer is ~41% of sales.
No clear moat yet.
The plant is a cost before it's a profit.
Why I'm long anyway:
In the middle of an AI boom, with capacity maxed out and chips forced out of China, I don't buy that TRT is cutting margins for no reason.
I think it's getting in cheap to land giants and earn the margin later.
High risk, high reward.
The test that settles it: next earnings in September.
Watch if revenue AND margin rise together.
Long $TRT. Not financial advice. DYOR.
✍1
This media is not supported in your browser
VIEW IN TELEGRAM
“Peter Thiel says that to be an entrepreneur you have to be contrarian—and right.”
“You have to be comfortable looking stupid for like a long time.”
“When I was calling those banks and saying: Hey we're a crypto company want to do this? They would hang up on me.”
“I'd go pitch the 30th venture investor and get a no.”
“We're willing to be misunderstood for a long time.”
“Entrenched interests will fight you.”
“Everything that's truly innovative and
breakthrough is going to upset an entrenched incumbent, eventually intersect with the government, and just
piss off some segment of the population who are like “How dare you question the status quo?”
— Brian Armstrong founder of Coinbase.
“You have to be comfortable looking stupid for like a long time.”
“When I was calling those banks and saying: Hey we're a crypto company want to do this? They would hang up on me.”
“I'd go pitch the 30th venture investor and get a no.”
“We're willing to be misunderstood for a long time.”
“Entrenched interests will fight you.”
“Everything that's truly innovative and
breakthrough is going to upset an entrenched incumbent, eventually intersect with the government, and just
piss off some segment of the population who are like “How dare you question the status quo?”
— Brian Armstrong founder of Coinbase.
🗿1
🇨🇭 Global One SI - Prompt
Photo
$ZSQR Appreciate seeing this play out.
Back in May, the AI infrastructure thesis around $ZSQR wasn’t the popular trade. The story needed time, execution, and market attention.
A lot of people were focused on the short-term volatility, but the underlying narrative remained intact.
Sometimes the market doesn’t reward the right thesis immediately.
It just rewards it eventually.
Congrats to everyone who stayed focused on the bigger picture.
Back in May, the AI infrastructure thesis around $ZSQR wasn’t the popular trade. The story needed time, execution, and market attention.
A lot of people were focused on the short-term volatility, but the underlying narrative remained intact.
Sometimes the market doesn’t reward the right thesis immediately.
It just rewards it eventually.
Congrats to everyone who stayed focused on the bigger picture.
💯1
🇨🇭 Global One SI - Prompt
My proposal for the transition to a post labor economy: Don’t wait until millions are displaced and then panic with welfare patches. Create an Automation Swap. Any company that replaces large amounts of human labor with AI or robots gets tax breaks, faster…
The real AI shock won’t be “robots take jobs.”
It’ll be when companies realize one weirdly capable person with agents can replace an entire department.
Do we share the productivity explosion, or do we let 12 people own the whole machine?
It’ll be when companies realize one weirdly capable person with agents can replace an entire department.
Do we share the productivity explosion, or do we let 12 people own the whole machine?
🔥1
This media is not supported in your browser
VIEW IN TELEGRAM
Me laughing with my new profitable homeboys after cutting off my unprofitable day ones
🐳1
🇨🇭 Global One SI - Prompt
🔌The AI economy runs on electricity. Every model. Every inference. Every data center. The grid was not built for this. Here's the stack of companies building the power and storage infrastructure the AI economy cannot function without. Thread. 🧵
🟠 Start with the anchor.
$GEV (GE Vernova) doesn't get talked about enough in frontier tech circles. It should.
Gas turbine backlog: 100 GW. Targeting 110 GW by year end.
Total backlog: $163 billion. Growing to $200 billion by 2027 -> a year ahead of schedule.
Q1 2026: Data center electrification orders hit $2.4 billion in a single quarter. More than all of 2025 combined.
FY2026 guidance: $44.5-$45.5 billion. Turbine slots sold out through 2030.
You cannot build an AI factory without power. You cannot get power at scale without $GEV
$GEV (GE Vernova) doesn't get talked about enough in frontier tech circles. It should.
Gas turbine backlog: 100 GW. Targeting 110 GW by year end.
Total backlog: $163 billion. Growing to $200 billion by 2027 -> a year ahead of schedule.
Q1 2026: Data center electrification orders hit $2.4 billion in a single quarter. More than all of 2025 combined.
FY2026 guidance: $44.5-$45.5 billion. Turbine slots sold out through 2030.
You cannot build an AI factory without power. You cannot get power at scale without $GEV
🇨🇭 Global One SI - Prompt
🔌The AI economy runs on electricity. Every model. Every inference. Every data center. The grid was not built for this. Here's the stack of companies building the power and storage infrastructure the AI economy cannot function without. Thread. 🧵
🟠 $BE (Bloom Energy) solved the hardest part of the AI power problem.
Time to power.
Hyperscalers can't wait 5 years for grid connections. Bloom's solid oxide fuel cells go behind the meter, bypass the grid entirely, and deliver power on-site.
Oracle understood this. Project Jupiter - one of the largest AI infrastructure campuses ever built - powered entirely by Bloom. Up to 2.45 GW.
Q1 2026: Revenue $751M. +130% YoY. Net income $70.7M. Profitable.
FY2026 guidance: $3.4-$3.8 billion. $20 billion backlog.
The grid has a queue. Bloom doesn't.
Time to power.
Hyperscalers can't wait 5 years for grid connections. Bloom's solid oxide fuel cells go behind the meter, bypass the grid entirely, and deliver power on-site.
Oracle understood this. Project Jupiter - one of the largest AI infrastructure campuses ever built - powered entirely by Bloom. Up to 2.45 GW.
Q1 2026: Revenue $751M. +130% YoY. Net income $70.7M. Profitable.
FY2026 guidance: $3.4-$3.8 billion. $20 billion backlog.
The grid has a queue. Bloom doesn't.
🇨🇭 Global One SI - Prompt
🔌The AI economy runs on electricity. Every model. Every inference. Every data center. The grid was not built for this. Here's the stack of companies building the power and storage infrastructure the AI economy cannot function without. Thread. 🧵
🟡The one nobody is talking about “clearly” enough.
$TE (T1 Energy). The only at-scale domestic solar manufacturer in America.
G1 Dallas: 5 GW operational. G2 Austin: 2.1 GW solar cell facility targeting first production Q4 2026.
Q1 2026: Revenue $177M. +232% YoY. Record adjusted EBITDA. Record net income from continuing operations.
900 MW supply agreement with Treaty Oak Clean Energy. $440M+ raised to advance the domestic supply chain.
Leopold Aschenbrenner's Situational Awareness LP bought $43.9 million of $TE shares.
Short reports on legitimate businesses are not risks. They are opportunities.
$TE (T1 Energy). The only at-scale domestic solar manufacturer in America.
G1 Dallas: 5 GW operational. G2 Austin: 2.1 GW solar cell facility targeting first production Q4 2026.
Q1 2026: Revenue $177M. +232% YoY. Record adjusted EBITDA. Record net income from continuing operations.
900 MW supply agreement with Treaty Oak Clean Energy. $440M+ raised to advance the domestic supply chain.
Leopold Aschenbrenner's Situational Awareness LP bought $43.9 million of $TE shares.
Short reports on legitimate businesses are not risks. They are opportunities.
🇨🇭 Global One SI - Prompt
🔌The AI economy runs on electricity. Every model. Every inference. Every data center. The grid was not built for this. Here's the stack of companies building the power and storage infrastructure the AI economy cannot function without. Thread. 🧵
🔵 The bridge between generation and storage.
$FLNC (Fluence Energy) just got placed at the center of AI infrastructure in a way the market hasn't fully priced.
NVIDIA. Siemens. Fluence.
Three-way collaboration to build the reference architecture for AI data center power stability. Fluence's SmartStack battery system is the power backbone of that blueprint.
$5.6 billion in signed backlog. $2 billion in new orders year to date — double last year's pace.
FY2026 guidance: $3.2-$3.6 billion.
When NVIDIA designs the AI factory of the future and puts your storage system in the blueprint, that is not a press release.
That is a moat.
$FLNC (Fluence Energy) just got placed at the center of AI infrastructure in a way the market hasn't fully priced.
NVIDIA. Siemens. Fluence.
Three-way collaboration to build the reference architecture for AI data center power stability. Fluence's SmartStack battery system is the power backbone of that blueprint.
$5.6 billion in signed backlog. $2 billion in new orders year to date — double last year's pace.
FY2026 guidance: $3.2-$3.6 billion.
When NVIDIA designs the AI factory of the future and puts your storage system in the blueprint, that is not a press release.
That is a moat.