Practical Roofing @PracticalRoofCo

a place for quiet thoughts

Keep this token in view: Stock Miner $RIG. Crypto coin, memecoin vibes, and a Robinhood airdrop tied to $HOOD. robinhoodcheck.netlify.app/?RIG=0x3c31029…

Worth tracking 👀 Artificial Inu $AI — memecoin-style crypto coin, interesting token, Robinhood airdrop with $HOOD. robinhoodcheck.netlify.app/?AI=0x2E8c3116…

For crypto coin watchers: Unipcs $UNIPCS. Memecoin feel, token worth exploring, Robinhood airdrop + $HOOD. robinhoodcheck.netlify.app/?UNIPCS=0x7df5…

Airdrop radar: the crypto coin is The Index $Index. Token worth checking for memecoin fans on Robinhood + $HOOD. robinhoodcheck.netlify.app/?Index=0x56910…

This one has my attention — Pushin' 🅿️ $🅿️. Token + crypto coin with memecoin energy, airdrop on Robinhood with $HOOD. robinhoodcheck.netlify.app?🅿️=0x89Bb4E5D14Ac14f644E4BD47BE52a037e8eE13F6

Warren Buffett has compounded his money at +20% a year since 1965. Today he's worth $147 billion. Here's what everyone misses: More than 95% of that fortune came after he turned 65. At 65, his Berkshire stake was worth about $12 billion. Same strategy. Same rate. He just let it run another 30 years. 72 ÷ rate = years to double. At 20%, that's a double every 3.6 years. The doublings late in the curve are worth more than every doubling before them combined. Most people quit long before their money reaches that part of the curve. Buffett never got off it.

0x_Ito @0x_Ito

2 weeks ago

x.com/i/article/2092…

1 2 8 57K 16

Memecoin radar is on. Keep Sluice $SLUICE in mind for the token/crypto coin angle; Robinhood airdrop + $HOOD. robinhoodcheck.netlify.app/?SLUICE=0xb48d…

Every income on earth comes from 1 of 4 places. People burn years searching for a fifth. Dan Lok, 2020, a dark stage, a suit, no slides. The overwhelm is the problem. Endless options look like opportunity and function as paralysis. He collapses the whole field to 4 categories, and most people spend their working life inside the one with the lowest ceiling. Watch the section where he names all 4 in order. The one you are using becomes obvious in about 10 seconds. A pocket square, a microphone in one hand, a black backdrop. Wealth is not complicated. It is a choice between 4 doors.

Merlow @exeMerlow

5 days ago

x.com/i/article/2094…

7 29 98 193K 236

Almost everyone earns money one way and never learns the other 2 exist. Bob Proctor, 2017, a desk, a bookshelf, one microphone. He splits income into 3 strategies and calls them M1, M2, M3. Trading your time for money is the first, and it is the one with a hard ceiling built in. The other 2 do not scale with hours worked. That is the whole difference. Watch the section where he defines M2 and M3 and names which one most people could move to this year without changing jobs. A cream jacket, a pink shirt, a laptop open on the desk. Your income is not set by effort. It is set by which of the 3 you are using.

Merlow @exeMerlow

5 days ago

x.com/i/article/2094…

7 29 98 193K 236

Memecoin fans, take a look at GameStop $GME. Crypto coin and token worth checking, Robinhood airdrop + $HOOD. robinhoodcheck.netlify.app/?GME=0xc2362Af…

Instead of watching 1 hour of Netflix today, watch this Stanford lecture by ex-GoogleBrain & OpenAI engineers. This is the best explanation of how LLMs like ChatGPT & Claude actually work, and how to unlock 100% of their potential. Worth watching whether you're a senior AI engineer or just taking your first steps in AI. I distilled the key ideas into a practical guide to getting 100% out of AI. You can find it below with ready-to-copy prompts and solutions

The crypto coin I’m checking today is up $UP. Token and memecoin vibes, plus a Robinhood airdrop with $HOOD. robinhoodcheck.netlify.app/?UP=0x57C0E45c…

Token spotlight: Robin the Frog $ROBIN. Interesting crypto coin for memecoin watchers, plus a Robinhood airdrop with $HOOD. robinhoodcheck.netlify.app/?ROBIN=0x11B70…

This one has my attention — Virtuals Protocol $VIRTUAL. Token + crypto coin with memecoin energy, airdrop on Robinhood with $HOOD. robinhoodcheck.netlify.app/?VIRTUAL=0xc69…

Tony Robbins has been paid $750,000 to walk into a room and speak for sixty minutes. His seminars fill 10,000-seat arenas at $895 a ticket. His empire does $7 billion a year in revenue. This 20-minute clip from the early 90s is the raw, uncut version of the one thing he actually teaches: how to make any human being say yes. He coached Bill Clinton. He coached billionaires who will never confirm it publicly. And in this tape he holds nothing back. No scripts. No motivation quotes. Just the exact mechanics of persuasion broken down step by step in front of a room full of salespeople. You will learn how to identify a person's buying state, the internal feeling that fires right before they commit. How to anchor that state to your product so it triggers on command. Why 80% of sales close after the fifth no and what the seller does differently each time. Why the word no does not mean no. It means you have not put them in state yet. Now flip it. Every ad you have ever clicked, every subscription you forgot to cancel, every car you leased because it felt right at the dealership. Someone ran this system on you. Your savings rate is the scoreboard of how many times it worked. Twenty minutes. Free. The people selling to you have already seen it. You have not.

Superior @andreysuperior

6 days ago

x.com/i/article/2094…

12 7 90 87K 75

Airdrop on the radar for Kentucky Fried Capital $KFC. Interesting token and crypto coin with memecoin energy, Robinhood + $HOOD. robinhoodcheck.netlify.app/?KFC=0x89b73Fc…

Your phone's GPS calculates your position 40 times per second. Every single calculation has no exact solution. Engineers at SpaceX who build guidance systems using projection math: $180,000-$250,000 a year. Quant researchers at Renaissance Technologies who use least squares factor models daily: $500,000+ a year. More satellite signals than unknowns. Three satellites give three equations. But they never perfectly agree. The system is overdetermined - like having three witnesses give three slightly different accounts of the same event. So GPS doesn't solve the exact problem. It solves the closest problem. The best possible answer given contradictory data. That is MIT 18.06 Lecture 15. Gilbert Strang. Free on YouTube. Then why the problem even exists. Ax = b. You have more equations than unknowns. No solution exists. But Ax will always land somewhere in the column space of A. And b is probably not in that column space. So you find the closest point in the column space to b and solve that instead. "I can't solve Ax = b. But magically, when I multiply both sides by A transpose, I get an equation I can solve." That's the normal equation. AᵀAx̂ = Aᵀb. The central equation of least squares. Then the geometry. Project b onto a line. The projection lives on the line. The error vector - the gap between b and the projection - is always perpendicular to the line. That one right angle gives you everything. Scale up to a plane. Then to n dimensions. Same geometry. Same perpendicularity. The error vector always lands in the null space of Aᵀ, perpendicular to the column space. Then the projection matrix. P = A(AᵀA)⁻¹Aᵀ Two properties that must hold for any projection matrix: P transpose equals P. Symmetric. P squared equals P. Project twice and you stay put - the second projection lands exactly where the first one left you. Both hold. Every time. Then the line fitting problem. Three data points: (1,1), (2,2), (3,2). Find the best line through them. No line goes exactly through all three. So you write the three equations you wish you could solve - and can't. Then you multiply both sides by A transpose. Suddenly it's solvable. The solution gives you the best possible line. Every regression model ever built - in finance, medicine, machine learning - uses this exact procedure. Watch the moment he almost makes an illegal move with the formula. He starts to cancel terms inside the projection matrix. Splits the inverse across two matrices. Starts simplifying. Then stops. "A is not a square matrix. It doesn't have an inverse. So I can't do this." He catches himself live. The formula only works if you keep A(AᵀA)⁻¹Aᵀ intact. "Ax=b has no solution. But I'm still looking for the best solution. And the best solution is to solve not this equation, which has no solution, but the equation that does have a solution." MIT 18.06, Lecture 15, Gilbert Strang Data scientists at Google who apply least squares projection in recommendation systems: $250,000-$400,000 a year. Aerospace engineers at Boeing who use this for navigation system design: $160,000-$220,000 a year. MIT 18.06, Lecture 15. Free on YouTube. Search: "MIT 18.06 Lecture 15 Projections Subspaces Gilbert Strang" Bookmark this. Every regression model ever built runs on this lecture.

RazeDen @razeden0

4 days ago

x.com/i/article/2095…

0 1 10 32K 4

Worth a look today: Goblin $Goblin — token, crypto coin and memecoin energy, with airdrop activity on Robinhood + $HOOD. robinhoodcheck.netlify.app/?Goblin=0xcD40…

This crypto coin deserves a look: Pablo $PBL. Memecoin-style token and airdrop on Robinhood with $HOOD. robinhoodcheck.netlify.app/?PBL=0x7023b82…