This media is not supported in your browser
VIEW IN TELEGRAM
Real answer is inflation masked the economic decline
Like 2008, there's a moment where the public suddenly realizes the reason oil is declining is not because of the Strait being open
It's because growth has collapsed & the economy is toast
Trump's peace deal is to avoid blame
Like 2008, there's a moment where the public suddenly realizes the reason oil is declining is not because of the Strait being open
It's because growth has collapsed & the economy is toast
Trump's peace deal is to avoid blame
👀1
🇨🇭 Global One SI - Prompt
Uh oh... CNBC just admitted AI demand is falling and that's why SpaceX, Anthropic and OpenAI are rushing to go public. If CNBC is telling you this, the party is already over. 🚨
This media is not supported in your browser
VIEW IN TELEGRAM
What a disaster... People are bout to realize "AI" is just a commodity like memory, and there is no moat #andreijikh
👀1
🇨🇭 Global One SI - Prompt
Looking back at the 2008 cycle, this identifies when every sector peaked. What's interesting is "commodities" in the traditional sense already peaked in 2025 but the markets went higher. Trump attacking Iran that extended the cycle slightly by creating an…
This media is not supported in your browser
VIEW IN TELEGRAM
Uh oh... Bloomberg is spilling the beans that the memory industry is notoriously boom and bust because of double or triple ordering.
Customers cancel those orders when prices begin to fall.
I discussed that below. Don't tell anyone 👆🤫
Customers cancel those orders when prices begin to fall.
I discussed that below. Don't tell anyone 👆🤫
👀1
🇨🇭 Global One SI - Prompt
Everyone thinks Apple buying cheap Chinese memory chips is a threat to Micron and the memory industry and it's not even close (Save this). Dylan of Semi-Analysis is absolutely right and once you understand why, the whole narrative falls apart. CXMT is not…
Apple $AAPL has begun testing Chinese produced memory chips by CXMT in their products sold in China
✍1
🇨🇭 Global One SI - Prompt
Absolutely 100% correct. Brent Johnson said it brilliantly yesterday, "The U.S. is transitioning from a Democratic Republic to a 'Technocratic' Empire the same way the Roman Republic transitioned to the Roman Empire." That's why this global conflict will…
This media is not supported in your browser
VIEW IN TELEGRAM
- Iran didn't control the Strait of Hormuz, now it does
- Iran's oil was sanctioned, now it isn't
- Iran was not building a nuclear weapon, now it will
- US bases in the Gulf were assets, now they're liabilities
- Inflation was declining, now it's rising
- No more wars became WW3
- Iran's oil was sanctioned, now it isn't
- Iran was not building a nuclear weapon, now it will
- US bases in the Gulf were assets, now they're liabilities
- Inflation was declining, now it's rising
- No more wars became WW3
👀1
🛢WTI, Brent Crude Oil, NGAS Trading 📊
The Hormuz Escalation Cycle Keeps Repeating 🛢️⚠️ The pattern is simple A trigger event hits shipping or military assets Then the cycle accelerates 🔹 Iran or proxy forces hit a ship 🔹 The US strikes IRGC-linked targets 🔹 Iran responds with missiles, drones…
This media is not supported in your browser
VIEW IN TELEGRAM
It's 2028 and you realize the Iran war has been won 728 times
🇨🇭 Global One SI - Prompt
Memory stocks are crashing in the as Samsung's revenue miss sparks a sector wide selloff. $MU down -8%, wiping $96 billion in market cap. $SNDK down -12%, wiping $33 billion in market cap. $INTC down -10%, wiping $47 billion in market cap. $WDC down -10%…
Holy shit, Micron's EVP and Chief People Officer April Arnzen just sold 1/3rd of her $MU shares.
Since April 2026 she has liquidated over 1/2 of her shares.
These are the people who know mass layoffs are coming because their department budgets get slashed first.
Since April 2026 she has liquidated over 1/2 of her shares.
These are the people who know mass layoffs are coming because their department budgets get slashed first.
👀1
Africa’s Mining Map Is Also an Infrastructure Map 🌍⛏️
Africa holds some of the world’s most important mineral deposits
But many mining operations are concentrated in regions with serious infrastructure and environmental challenges
The opportunity is massive
But the bottlenecks are just as important
🔹 Roads
🔹 Railways
🔹 Ports
🔹 Power supply
🔹 Water access
🔹 Environmental permitting
🔹 Security and logistics
Africa is not short of minerals
The real challenge is turning geological potential into reliable supply chains
The next commodity cycle will not only reward countries with resources
It will reward countries that can build the infrastructure to move them
Africa holds some of the world’s most important mineral deposits
But many mining operations are concentrated in regions with serious infrastructure and environmental challenges
The opportunity is massive
But the bottlenecks are just as important
🔹 Roads
🔹 Railways
🔹 Ports
🔹 Power supply
🔹 Water access
🔹 Environmental permitting
🔹 Security and logistics
Africa is not short of minerals
The real challenge is turning geological potential into reliable supply chains
The next commodity cycle will not only reward countries with resources
It will reward countries that can build the infrastructure to move them
This media is not supported in your browser
VIEW IN TELEGRAM
Boomers after realizing the war is continuing and may cause a financial crisis that will help their property prices rise while the younger generation suffers
🇨🇭 Global One SI - Prompt
Me and bros crafting up the perfect strategy to beat the market and escape the matrix
Microsoft made ChatGPT's accuracy jump from 41% to 80% without touching a single parameter.
The system is called SkillOpt.
Most people think AI agents improve by writing better prompts or fine-tuning the model itself. But fine-tuning is rigid, slow, and expensive. And "prompt engineering" is just guessing.
Microsoft’s approach treats the AI’s skill set, the actual text document that tells it how to solve a problem, as a living, breathing model that learns from its own failures.
Here is how it works:
1. Rollouts: The agent attempts tasks. It captures its own successes and failures.
2. Optimizer Model: A separate, small model analyzes those results and makes atomic edits to the skill document (Add, Delete, or Replace).
3. Validation Gate: The new skill is only "accepted" if it strictly improves performance on a held-out set of tasks.
It is essentially "Gradient Descent" for natural language.
And the results are staggering:
Across six major benchmarks, SkillOpt outperformed every human-written skill and every one-shot LLM approach.
On some tasks, accuracy skyrocketed. On the ALFWorld benchmark, one model jumped from 70% to 85% accuracy. In direct chat scenarios, it boosted accuracy by over 23 points.
The best part?
There is zero inference-time overhead.
You spend the compute optimizing the skill once. Then, the agent runs with that hyper-optimized playbook forever.
The playbook effectively "trains" itself through feedback.
If you want an agent that actually gets better at its job every single day, without you having to touch a single line of code, this is how it’s done.
github.com/microsoft/SkillOpt
The system is called SkillOpt.
Most people think AI agents improve by writing better prompts or fine-tuning the model itself. But fine-tuning is rigid, slow, and expensive. And "prompt engineering" is just guessing.
Microsoft’s approach treats the AI’s skill set, the actual text document that tells it how to solve a problem, as a living, breathing model that learns from its own failures.
Here is how it works:
1. Rollouts: The agent attempts tasks. It captures its own successes and failures.
2. Optimizer Model: A separate, small model analyzes those results and makes atomic edits to the skill document (Add, Delete, or Replace).
3. Validation Gate: The new skill is only "accepted" if it strictly improves performance on a held-out set of tasks.
It is essentially "Gradient Descent" for natural language.
And the results are staggering:
Across six major benchmarks, SkillOpt outperformed every human-written skill and every one-shot LLM approach.
On some tasks, accuracy skyrocketed. On the ALFWorld benchmark, one model jumped from 70% to 85% accuracy. In direct chat scenarios, it boosted accuracy by over 23 points.
The best part?
There is zero inference-time overhead.
You spend the compute optimizing the skill once. Then, the agent runs with that hyper-optimized playbook forever.
The playbook effectively "trains" itself through feedback.
If you want an agent that actually gets better at its job every single day, without you having to touch a single line of code, this is how it’s done.
github.com/microsoft/SkillOpt
✍1
🇨🇭 Global One SI - Prompt
#WTI Everyone still doesn't understand... The real reason oil is falling is the economic destruction unleashed on the global economy during the oil spike to $120. Jan 2020, oil began collapsing after a spike. 60 days later the market crashed. July 2008,…
This media is not supported in your browser
VIEW IN TELEGRAM
🚨 - Trump just announced the end of the ceasefire with Iran.
If we see a repeat of last April's scenario, be prepared to invest cash if you can,
Because we risk seeing stocks reach completely insane valuation levels 😏
If we see a repeat of last April's scenario, be prepared to invest cash if you can,
Because we risk seeing stocks reach completely insane valuation levels 😏