Offshore
Video
Chips & SaaS
RT @JanJekielek: “The American people are the last bulwark against CCP hegemony.”

“Thr more divided we are, the harder it is for us to resist the CCP’s ultimate political objective.”

“Which is global hegemony.”

Former NSC Chief of Staff @alexgrayforok told me that China uses extreme partisan groups to sow division in the United States.

“The Soviets made a very professional effort to sow discord in the United States.”

“The Chinese have inherited those lessons, and in many ways, they’ve refined them, and they’ve made them more sophisticated and more effective.”

“The goal is to use organizations that are partisan, that are extreme, that sow disunity and disharmony within American society to create wedges that make us vulnerable to the CCP’s division.”

“If they can divide us, confuse us, keep us at each other’s throats, they will have an easier path towards their ultimate goal.”
tweet
Offshore
Video
Moon Dev
Stop Building Slop: 3 AI Agent Use Cases That Actually Front-Run Wall Street

most people are using the most powerful automation tools in the world to generate content slop while the real players are using them to hunt for money in places wall street hasn't even noticed yet. we are currently living through a gold rush where the pickaxes are ai agents but almost everyone is digging in the wrong spot. if you want to know how to turn these tools into a literal money printer you have to understand why most of the current use cases are actually a total waste of time

i go by the name moon dev and i believe that code is the great equalizer in a world designed to liquidate the average person. before i learned to code i spent hundreds of thousands of dollars on developers for apps because i thought i wasn't smart enough to do it myself. it was a massive mistake that cost me a fortune in over trading and liquidations until i realized that i had to automate my survival

the primary problem with tools like openclaw right now is that the average person is using them for tasks that don't move the needle. they are out here building agents to summarize reddit threads or create daily youtube digests which is just more noise in an already loud world. why would you spend your computing credits on organizing content when you could be using them to find a social arbitrage opportunity before it hits the mainstream news

one of the few use cases that actually makes sense involves turning an agent into a tireless tiktok scroller to find trends before they blow up. imagine an agent that sits there all day analyzing sentiment and spotting products or stocks that are starting to go viral in specific niche communities. this is how you find trends before wall street knows they exist because you are essentially front running the cultural consciousness

there is a massive trap waiting for anyone who jumps into automation without knowing how to actually build their own environment. i decided to learn to code live on youtube to show people that you don't need to be a genius to build fully automated systems that trade for you. without that foundation you are just a passenger in someone else's car hoping they don't drive you off a cliff

as an algorithmic trader i follow a very specific process called rbi which stands for research backtest and implement. most of the openclaw use cases i see skip the research and backtesting parts and go straight to implementation which is just gambling with extra steps. if your agent isn't helping you find a statistical edge that you can prove with data then you are just burning electricity for no reason

a lot of people get excited about agents that can manage their dms or declutter their inbox because it feels like they are being productive. while that might save you a few minutes a day it doesn't actually solve the bottleneck of creating real wealth in the markets. you need to be looking for the use cases that scale your brain instead of just cleaning your digital house

the real secret to winning this game is figuring out how to create a million versions of yourself working all day on your craft. i keep a whiteboard behind me that says i want to become the best in the world at machine learning so i can have a million of me running at once. that sounds big headed to some but in the world of ai it is a literal possibility if you can architect the system correctly

we are constantly looking for a specific kind of arbitrage where the cost of the automation is lower than the profit it generates. for example if i pay two hundred dollars for a cloud subscription i am looking for a use case that prints at least two hundred and ten dollars back. until you find that ten dollar profit margin you are just a customer of the ai companies rather than a partner in the new economy

i often feel like gilfoyle from silicon valley because i look at these shiny new ui tools and wonder why i wouldn't just run it in a terminal. for a developer it is often a [...]
Offshore
Moon Dev Stop Building Slop: 3 AI Agent Use Cases That Actually Front-Run Wall Street most people are using the most powerful automation tools in the world to generate content slop while the real players are using them to hunt for money in places wall street…
hundred times faster and easier to just write a python script than to set up a complex autonomous agent. you have to be careful not to fall in love with the technology itself and lose sight of the actual goal which is printing money

there is a fire idea involving a trading swarm coordinator where multiple agents share signals and stack conviction on a single move. instead of three separate bots trading blindly they could communicate in real time so that a whale watcher agent can alert a risk agent. this creates a unified trading brain that reacts in seconds to market shifts that a human would miss while they are sleeping

every day i test these new technologies to see if they can finally beat the systems i have already built in the terminal. it is a constant battle between the simplicity of a well written script and the potential of an autonomous agent that can navigate the web. the moment i find an agent that can consistently find and execute a new edge i will scale it to a level that most people can't even imagine

the transition from being a manual trader to a fully automated developer is the hardest thing you will ever do but it is also the most rewarding. i spent years being scared of brackets and semicolons until i realized they were the keys to my freedom from the charts. you have to be willing to iterate through hundreds of failed bots until you find the one that finally starts to work

right now i have multiple agents running across different cloud instances all working on separate parts of my trading empire. some are focused on polymarket while others are hunting for social arbitrage on x and tiktok for my e commerce ideas. this is the only way to stay competitive in a market where the bots are already taking over everything

if you are just starting out you should focus on building something that is not doable in a simple terminal environment. if your idea can be done with a basic python script then don't waste your time and credits setting up an autonomous agent. the value of these agents is in their ability to handle the messy and unpredictable parts of the internet that code usually struggles with

i challenge you to stop looking for the easy way out and actually put in the hours to beat on your craft until you are the best. the talent might be something you are born with but the skill is only developed through thousands of hours of building and failing. don't settle for the slop that everyone else is building because the real money is found in the deep end where most people are too afraid to swim

at the end of the day it doesn't matter how cool your tech stack is if it isn't actually moving you toward your vision. you have to create your own environment and realize that what you can imagine you can eventually create if you stay persistent. the machines are here to stay and you either learn to command them or you get left behind by the people who do
tweet
Offshore
Video
Startup Archive
Peter Thiel on why picking a job because it will “look good on your resume” is bad idea

“What you end up doing when you think of the future as indefinite is you pick a job that will be good on your resume because it will then lead to a different job later on.”

Instead of viewing the future as indefinite, Thiel suggests that the students in the audience think of the future on three horizons: short term, medium term, and long term. And he defines them as follows:

1. Short term = Are you going to be happy and will you be learning a lot?

2. Medium term = How good is it on your resume?

3. Long term = Are you working on something important or meaningful?

“I think we are in a skewed world where we overweight the medium term, and we underweight some combination of the short term and the long term.”

Thiel recommends recalibrating it to:

“Focus on some combination of short term and long term. And forget about the medium term.”
tweet
Offshore
Photo
Bourbon Capital
RT @BourbonCap: Stocks to Buy and Hold for the Next Decade (Part 2)

1. $SOFI - SoFi Technologies

SoFi has evolved from a single-product lender into a diversified digital financial ecosystem with bank-level economics, software-like margins, and expanding fintech infrastructure capabilities.

The company operates as a modern, mobile-first financial platform targeting younger consumers across personal banking, lending, investing, and crypto.

Product adoption continues to accelerate, with 1.6 million new products added in Q4, bringing the total to over 20 million products, up 37% YoY.

Member growth remains a key driver of long-term expansion. In Q4, SoFi added a record 1 million new members, bringing total membership to 13.7 million, up 35% YoY. This marked the first quarter in the company’s history with more than one million net additions.

SoFi has reached an important inflection point. The company has delivered positive net income since 2024, while Adjusted EBITDA reached $318 million in Q4 (+60% YoY), representing a 31% margin, exceeding management’s long-term 30% target. Full-year EBITDA reached $1.1 billion (+58% YoY) at a 29% margin, reinforcing SoFi’s transition from growth-at-all-costs to durable profitability.
tweet
Offshore
Photo
The Transcript
RT @TheTranscript_: $SPG Simon Property Group CEO: "Hopefully, AI will solve it for us, so we don’t have to negotiate. It will just say: ‘Here is the rent that the tenant and landlord should agree on.’”" https://t.co/7JzKwDpFC0
tweet
Offshore
Video
Dimitry Nakhla | Babylon Capital®
RT @DimitryNakhla: Bill Ackman on Psychology of Investing 🧠

“You have to get excited when things get cheaper and get concerned when things get more expensive… Stocks can trade at any price in the short term. If you know what a business is worth and understand it extremely well, volatility bothers you much less…

It’s a combination of being personally secure and knowing what you own. Over time, you build calluses… I’m a pretty emotional person — but not in investing.”

𝐓𝐡𝐞 𝐥𝐞𝐬𝐬𝐨𝐧: 𝘛𝘦𝘮𝘱𝘦𝘳𝘢𝘮𝘦𝘯𝘵 𝘤𝘰𝘶𝘱𝘭𝘦𝘥 𝘸𝘪𝘵𝘩 𝘢 𝘥𝘦𝘦𝘱 𝘶𝘯𝘥𝘦𝘳𝘴𝘵𝘢𝘯𝘥𝘪𝘯𝘨 𝘰𝘧 𝘵𝘩𝘦 𝘣𝘶𝘴𝘪𝘯𝘦𝘴𝘴𝘦𝘴 𝘺𝘰𝘶 𝘪𝘯𝘷𝘦𝘴𝘵 𝘪𝘯 𝘪𝘴 𝘢 𝘤𝘰𝘮𝘱𝘦𝘵𝘪𝘵𝘪𝘷𝘦 𝘢𝘥𝘷𝘢𝘯𝘵𝘢𝘨𝘦.
___

1️⃣ 𝐏𝐨𝐬𝐢𝐭𝐢𝐨𝐧 𝐬𝐢𝐳𝐢𝐧𝐠 𝐚𝐧𝐝 𝐬𝐮𝐫𝐯𝐢𝐯𝐚𝐥 𝐩𝐬𝐲𝐜𝐡𝐨𝐥𝐨𝐠𝐲

Large drawdowns don’t just test portfolios — they test 𝘣𝘪𝘰𝘭𝘰𝘨𝘺.

Sharp declines can trigger the same fight-or-flight response governed by the 𝘢𝘮𝘺𝘨𝘥𝘢𝘭𝘢.

When that switch flips, decision-making shifts from logic → survival.

𝐀𝐜𝐤𝐦𝐚𝐧’𝐬 𝐢𝐧𝐬𝐢𝐠𝐡𝐭: 𝘐𝘯𝘷𝘦𝘴𝘵 𝘢𝘮𝘰𝘶𝘯𝘵𝘴 𝘵𝘩𝘢𝘵 𝘢𝘭𝘭𝘰𝘸 𝘺𝘰𝘶 𝘵𝘰 𝘳𝘦𝘮𝘢𝘪𝘯 𝘱𝘴𝘺𝘤𝘩𝘰𝘭𝘰𝘨𝘪𝘤𝘢𝘭𝘭𝘺 𝘢𝘯𝘥 𝘧𝘪𝘯𝘢𝘯𝘤𝘪𝘢𝘭𝘭𝘺 𝘤𝘰𝘮𝘧𝘰𝘳𝘵𝘢𝘣𝘭𝘦. 𝘞𝘩𝘦𝘯 𝘴𝘶𝘳𝘷𝘪𝘷𝘢𝘭 𝘪𝘯𝘴𝘵𝘪𝘯𝘤𝘵𝘴 𝘢𝘳𝘦𝘯’𝘵 𝘢𝘤𝘵𝘪𝘷𝘢𝘵𝘦𝘥, 𝘳𝘢𝘵𝘪𝘰𝘯𝘢𝘭𝘪𝘵𝘺 𝘴𝘶𝘳𝘷𝘪𝘷𝘦𝘴 𝘷𝘰𝘭𝘢𝘵𝘪𝘭𝘪𝘵𝘺.
___

2️⃣ 𝐂𝐨𝐧𝐯𝐢𝐜𝐭𝐢𝐨𝐧 𝐜𝐨𝐦𝐞𝐬 𝐟𝐫𝐨𝐦 𝐮𝐧𝐝𝐞𝐫𝐬𝐭𝐚𝐧𝐝𝐢𝐧𝐠

Volatility feels very different when you deeply understand:

What the business is worth

The fundamental drivers of value

The durability of its moat

𝘞𝘩𝘦𝘯 𝘱𝘳𝘪𝘤𝘦 𝘧𝘢𝘭𝘭𝘴 𝘣𝘶𝘵 𝘪𝘯𝘵𝘳𝘪𝘯𝘴𝘪𝘤 𝘷𝘢𝘭𝘶𝘦 𝘳𝘦𝘮𝘢𝘪𝘯𝘴 𝘪𝘯𝘵𝘢𝘤𝘵, 𝘥𝘪𝘴𝘤𝘰𝘮𝘧𝘰𝘳𝘵 𝘤𝘢𝘯 𝘵𝘳𝘢𝘯𝘴𝘧𝘰𝘳𝘮 𝘪𝘯𝘵𝘰 𝘰𝘱𝘱𝘰𝘳𝘵𝘶𝘯𝘪𝘵𝘺. 𝘐𝘧 𝘺𝘰𝘶 𝘬𝘯𝘰𝘸 𝘸𝘩𝘢𝘵 𝘺𝘰𝘶 𝘰𝘸𝘯, 𝘥𝘦𝘤𝘭𝘪𝘯𝘦𝘴 𝘣𝘰𝘵𝘩𝘦𝘳 𝘺𝘰𝘶 𝘭𝘦𝘴𝘴. 𝘖𝘧𝘵𝘦𝘯, 𝘵𝘩𝘦𝘺 𝘣𝘦𝘤𝘰𝘮𝘦 𝘮𝘰𝘮𝘦𝘯𝘵𝘴 𝘰𝘧 𝘦𝘹𝘤𝘪𝘵𝘦𝘮𝘦𝘯𝘵 — 𝘵𝘩𝘦 𝘤𝘩𝘢𝘯𝘤𝘦 𝘵𝘰 𝘢𝘤𝘤𝘶𝘮𝘶𝘭𝘢𝘵𝘦 𝘮𝘰𝘳𝘦 𝘢𝘵 𝘣𝘦𝘵𝘵𝘦𝘳 𝘱𝘳𝘪𝘤𝘦𝘴.
___

3️⃣ 𝐊𝐧𝐨𝐰 𝐲𝐨𝐮𝐫𝐬𝐞𝐥𝐟

Ackman openly acknowledges being emotional. That honesty is critical.

“𝘛𝘩𝘦 𝘧𝘪𝘳𝘴𝘵 𝘱𝘳𝘪𝘯𝘤𝘪𝘱𝘭𝘦 𝘪𝘴 𝘵𝘩𝘢𝘵 𝘺𝘰𝘶 𝘮𝘶𝘴𝘵 𝘯𝘰𝘵 𝘧𝘰𝘰𝘭 𝘺𝘰𝘶𝘳𝘴𝘦𝘭𝘧 — 𝘢𝘯𝘥 𝘺𝘰𝘶 𝘢𝘳𝘦 𝘵𝘩𝘦 𝘦𝘢𝘴𝘪𝘦𝘴𝘵 𝘱𝘦𝘳𝘴𝘰𝘯 𝘵𝘰 𝘧𝘰𝘰𝘭.” — Richard Feynman

We’re all human.

𝐅𝐞𝐞𝐥𝐢𝐧𝐠 𝐞𝐦𝐨𝐭𝐢𝐨𝐧 𝐢𝐬 𝙣𝙤𝙧𝙢𝙖𝙡.

𝐀𝐜𝐭𝐢𝐧𝐠 𝐢𝐦𝐩𝐮𝐥𝐬𝐢𝐯𝐞𝐥𝐲 𝐢𝐬 𝙤𝙥𝙩𝙞𝙤𝙣𝙖𝙡.

Great investors build systems:

Preset rules

Allocation frameworks

Behavioral guardrails

Over time, experience itself becomes conditioning.

Just as @davidgoggins speaks about 𝙘𝙖𝙡𝙡𝙤𝙪𝙨𝙞𝙣𝙜 𝙩𝙝𝙚 𝙢𝙞𝙣𝙙 𝙩𝙝𝙧𝙤𝙪𝙜𝙝 𝙧𝙚𝙥𝙚𝙖𝙩𝙚𝙙 𝙨𝙩𝙧𝙚𝙨𝙨, 𝙞𝙣𝙫𝙚𝙨𝙩𝙤𝙧𝙨 𝙗𝙪𝙞𝙡𝙙 𝙥𝙨𝙮𝙘𝙝𝙤𝙡𝙤𝙜𝙞𝙘𝙖𝙡 𝙧𝙚𝙨𝙞𝙡𝙞𝙚𝙣𝙘𝙚 𝙩𝙝𝙧𝙤𝙪𝙜𝙝 𝙧𝙚𝙥𝙚𝙖𝙩𝙚𝙙 𝙚𝙭𝙥𝙤𝙨𝙪𝙧𝙚 𝙩𝙤 𝙫𝙤𝙡𝙖𝙩𝙞𝙡𝙞𝙩𝙮.

You develop “mental calluses.”

𝙑𝙤𝙡𝙖𝙩𝙞𝙡𝙞𝙩𝙮 𝙨𝙩𝙤𝙥𝙨 𝙛𝙚𝙚𝙡𝙞𝙣𝙜 𝙡𝙞𝙠𝙚 𝙙𝙖𝙣𝙜𝙚𝙧 — 𝙖𝙣𝙙 𝙨𝙩𝙖𝙧𝙩𝙨 𝙛𝙚𝙚𝙡𝙞𝙣𝙜 𝙡𝙞𝙠𝙚 𝙥𝙖𝙧𝙩 𝙤𝙛 𝙩𝙝𝙚 𝙥𝙧𝙤𝙘𝙚𝙨𝙨.
___

Video: Lex Fridman Poscast (02/20/2024) @BillAckman
tweet
Offshore
Photo
Illiquid
🤔 anyway I got the idea from someone buying theme parks so I guess nothing is really original. https://t.co/yvHOdEDTun

$dis is entering AI age of abundance with valuable in-person experiences (cruises, theme parks), valuable IP, and declining production costs (video models) at 15 p/e.
- Illiquid
tweet
Michael Fritzell (Asian Century Stocks)
RT @GuastyWinds: After 2 awesome years in the US I’ve decided to move back home to pursue an exciting opty. outside of public markets (will share more at some point). On the side I’ll still be trading and sharing ideas - stay tuned for a big one coming next week!
tweet
Offshore
Photo
Benjamin Hernandez😎
China’s probe into its top military general jolted markets, raising uncertainty around leadership stability. Geopolitical risk could drive volatility in defense and China-linked stocks.

$LMT $NOC $FXI https://t.co/JkuE0mAgCR
tweet
Michael Fritzell (Asian Century Stocks)
RT @jptissot1: https://t.co/WrgU6oIUfx
tweet