Offshore
Photo
The Few Bets That Matter
RT @WealthyReadings: Investing is unforgiving. One small mistake can be very costly, and there’s no way back.
$NBIS might be that mistake for me this year.
This isn’t hindsight bias. I said the very next day that selling was a mistake as the stock didn't break my conditions to hold. Sentiment took over and I took a decision I shouldn’t have.
That’s all it takes.
The last week and a half cost me a few salaries in unrealized P&L. One emotinal afternoon cost me months of work.
I’m not here to complain. I’m here to be transparent, to illustrate how critical systems - and respecting them, really are.
Your system exists to protect you from yourself.
Ideas, opinions, convictions can make money. But they can't regularly outperform. Over time, convictions turn into bias, and bias costs.
Systems only can compound over decades. And the single most important rule is simple: don’t break it.
I broke mine. Now I have to work on fixing that.
Mistakes are opportunities to improve.
tweet
RT @WealthyReadings: Investing is unforgiving. One small mistake can be very costly, and there’s no way back.
$NBIS might be that mistake for me this year.
This isn’t hindsight bias. I said the very next day that selling was a mistake as the stock didn't break my conditions to hold. Sentiment took over and I took a decision I shouldn’t have.
That’s all it takes.
The last week and a half cost me a few salaries in unrealized P&L. One emotinal afternoon cost me months of work.
I’m not here to complain. I’m here to be transparent, to illustrate how critical systems - and respecting them, really are.
Your system exists to protect you from yourself.
Ideas, opinions, convictions can make money. But they can't regularly outperform. Over time, convictions turn into bias, and bias costs.
Systems only can compound over decades. And the single most important rule is simple: don’t break it.
I broke mine. Now I have to work on fixing that.
Mistakes are opportunities to improve.
Few $NBIS notes after this quarter.
I'll be the bear, once more.
I continue to believe the market will punish the stock - or not reward it as much as many expect.
Not because the company isn’t excellent, but because it did not reward $GOOG, so why would it reward $NBIS for the same behavior?
Fundamentally, everyone will be bullish. Demand is through the roof, compute was sold out, management is planning to build more sites, etc...
Everything FinX wants to see.
From a market perspective, Q4 CapEx slowed down, guidance talks about ~20% increase of contracted power for FY26 without news on connected power, except for the upgrade from 7 sites to 16 sites.
This means FY26 CapEx will accelerate - just like for everyone else, and won't slow down FY27 as contracted power continues to climb.
More spending. Which was punished across all hyperscalers.
Also note that ARR guidance wasn’t increased, meaning no beat expected hence nothing above expectations and no buildouts closing faster than expected.
Some will say "why would you want more? It doesn't matter, they are executing at their pace"
I disagree. Acceleration is everything, otherwise you'll miss on expectations just like they did.
That revenue miss is due to real-world constraints, as I’ve shared yesterday and for months: you cannot build faster than physics and logistics allow you to.
The issue is that growth factually slows/doesn't accelerate. Growth stocks work on acceleration not stable growth.
The why doesn’t matter, even if you’re supply constrained.
Growth slows, CapEx increases, cash generation decreases, and there are no certainties that demand won’t be fulfilled by other hyperscalers by the time infrastructure is built.
Like many of you, I believe there will be demand and everything will be fine. But today, you cannot know. You can bet on it, but you cannot know.
That is the issue. And that is why the market might react like it did for $GOOG.
I continue to believe the company is excellent and its future is bright. And that the stock won’t be rewarded as much as many expect in the short term.
I’d love to be wrong. tweet
The Transcript
RT @TheTranscript_: $RBLX CEO: "Every day, we capture roughly 30,000 years of human interaction data on Roblox in a PII and privacy compliant way. We're actively using this data to develop and train AI models that continue to bring our vision to life. I want to highlight that we're internally now running over 400 AI models. "
tweet
RT @TheTranscript_: $RBLX CEO: "Every day, we capture roughly 30,000 years of human interaction data on Roblox in a PII and privacy compliant way. We're actively using this data to develop and train AI models that continue to bring our vision to life. I want to highlight that we're internally now running over 400 AI models. "
tweet
Offshore
Video
memenodes
Who got you into crypto?
Thanks to them you’re going to die 10 years younger from stress and bad habits https://t.co/jkBnOI0EP1
tweet
Who got you into crypto?
Thanks to them you’re going to die 10 years younger from stress and bad habits https://t.co/jkBnOI0EP1
tweet
Offshore
Video
memenodes
I can see why they used up all the condoms in the olympic village https://t.co/lRcnJaByRz
tweet
I can see why they used up all the condoms in the olympic village https://t.co/lRcnJaByRz
tweet
Offshore
Video
memenodes
RT @naiivememe: the refrigerator since 2018 watching me turn off the TV to let it rests
https://t.co/tJH4DFN3KP
tweet
RT @naiivememe: the refrigerator since 2018 watching me turn off the TV to let it rests
https://t.co/tJH4DFN3KP
tweet
Javier Blas
"Global oil prices have not declined despite a global surplus," says Goldman Sachs. "One key reason for this disconnect is that much of this global surplus has materialized as rising inventories of sanctioned crude "stuck at sea" while OECD stocks in pricing centers have remained stable."
tweet
"Global oil prices have not declined despite a global surplus," says Goldman Sachs. "One key reason for this disconnect is that much of this global surplus has materialized as rising inventories of sanctioned crude "stuck at sea" while OECD stocks in pricing centers have remained stable."
tweet
Offshore
Photo
God of Prompt
🚨 Holy shit… Google just published one of the cleanest demonstrations of real multi-agent intelligence I’ve seen so far.
Not another “look, two chatbots are talking” demo.
An actual framework for how agents can infer who they’re interacting with and adapt on the fly.
The paper is “Multi-agent cooperation through in-context co-player inference.”
The core idea is deceptively simple:
In multi-agent environments, performance doesn’t just depend on the task.
It depends on who you’re paired with.
Most current systems ignore this.
They optimize against an average opponent.
Or assume fixed partner behavior.
Or hard-code roles.
Google does something smarter.
They let the model infer its co-player’s strategy directly from the interaction history inside the context window.
No retraining, separate belief model and no explicit opponent classifier.
Just in-context inference.
The agent observes a few rounds of behavior. Forms an implicit hypothesis about its partner’s type. Then updates its own strategy accordingly.
This turns static policies into adaptive ones.
The experiments are structured around cooperative and social dilemma games where partner types vary:
Some partners are fully cooperative.
Some are selfish.
Some are stochastic.
Some strategically defect.
Agents without co-player inference treat all partners the same.
Agents with inference adjust.
And the performance gap is significant.
What makes this paper uncomfortable for a lot of current “multi-agent” hype is how clearly it shows what real coordination requires.
First, coordination is not just communication. It’s modeling the incentives and likely actions of others.
Second, robustness matters. An agent that cooperates blindly gets exploited. An agent that defects blindly loses cooperative gains. The system must dynamically balance trust and caution.
Third, adaptation must happen at inference time. In real deployments, you cannot retrain every time the population changes.
The most interesting part is that this capability emerges purely from structured context.
The model isn’t fine-tuned to classify opponent types explicitly. It uses behavioral traces embedded in the prompt to infer latent strategy.
That’s belief modeling through language.
And it scales.
Think about where this matters outside toy games:
Autonomous trading systems reacting to different market participants.
Negotiation agents interacting with unpredictable humans.
Distributed AI workflows coordinating across departments.
Swarm robotics where teammate reliability varies.
In all these settings, static competence is not enough.
Strategic awareness is the bottleneck.
The deeper shift is philosophical.
We’ve been treating LLM agents as isolated optimizers.
This paper moves us toward agents that reason about other agents reasoning about them.
That’s recursive modeling.
And once that loop becomes stable, you no longer have “a chatbot.”
You have a participant in a strategic ecosystem.
The takeaway isn’t that multi-agent AI is solved.
It’s that most current systems aren’t even attempting the hard part.
Real multi-agent intelligence isn’t multiple prompts in parallel.
It’s adaptive belief formation under uncertainty.
And this paper is one of the first clean proofs that large models can do that using nothing but context.
Paper: Multi-agent cooperation through in-context co-player inference
tweet
🚨 Holy shit… Google just published one of the cleanest demonstrations of real multi-agent intelligence I’ve seen so far.
Not another “look, two chatbots are talking” demo.
An actual framework for how agents can infer who they’re interacting with and adapt on the fly.
The paper is “Multi-agent cooperation through in-context co-player inference.”
The core idea is deceptively simple:
In multi-agent environments, performance doesn’t just depend on the task.
It depends on who you’re paired with.
Most current systems ignore this.
They optimize against an average opponent.
Or assume fixed partner behavior.
Or hard-code roles.
Google does something smarter.
They let the model infer its co-player’s strategy directly from the interaction history inside the context window.
No retraining, separate belief model and no explicit opponent classifier.
Just in-context inference.
The agent observes a few rounds of behavior. Forms an implicit hypothesis about its partner’s type. Then updates its own strategy accordingly.
This turns static policies into adaptive ones.
The experiments are structured around cooperative and social dilemma games where partner types vary:
Some partners are fully cooperative.
Some are selfish.
Some are stochastic.
Some strategically defect.
Agents without co-player inference treat all partners the same.
Agents with inference adjust.
And the performance gap is significant.
What makes this paper uncomfortable for a lot of current “multi-agent” hype is how clearly it shows what real coordination requires.
First, coordination is not just communication. It’s modeling the incentives and likely actions of others.
Second, robustness matters. An agent that cooperates blindly gets exploited. An agent that defects blindly loses cooperative gains. The system must dynamically balance trust and caution.
Third, adaptation must happen at inference time. In real deployments, you cannot retrain every time the population changes.
The most interesting part is that this capability emerges purely from structured context.
The model isn’t fine-tuned to classify opponent types explicitly. It uses behavioral traces embedded in the prompt to infer latent strategy.
That’s belief modeling through language.
And it scales.
Think about where this matters outside toy games:
Autonomous trading systems reacting to different market participants.
Negotiation agents interacting with unpredictable humans.
Distributed AI workflows coordinating across departments.
Swarm robotics where teammate reliability varies.
In all these settings, static competence is not enough.
Strategic awareness is the bottleneck.
The deeper shift is philosophical.
We’ve been treating LLM agents as isolated optimizers.
This paper moves us toward agents that reason about other agents reasoning about them.
That’s recursive modeling.
And once that loop becomes stable, you no longer have “a chatbot.”
You have a participant in a strategic ecosystem.
The takeaway isn’t that multi-agent AI is solved.
It’s that most current systems aren’t even attempting the hard part.
Real multi-agent intelligence isn’t multiple prompts in parallel.
It’s adaptive belief formation under uncertainty.
And this paper is one of the first clean proofs that large models can do that using nothing but context.
Paper: Multi-agent cooperation through in-context co-player inference
tweet