Skulduggery

[Divyanshi @Divyyanshishrma]

My friend makes $1.2 million a year as an Anthropic engineer.

I asked him how he learned prompting so well.

He sent me a video that was never supposed to get out. Their core team's prompting playbook.

You won’t find anything better about prompting than this video.

I watched it last night.

Halfway through, I realized I've been using Claude completely wrong for two years.

Watch it and save!

[SONIA @S0N_IA]

Ray Dalio acaba de dar una clase de 42 minutos que vale más que muchos cursos de inversión de miles de dólares. Y es GRATIS.

Olvida las fórmulas mágicas para hacerte rico de la noche a la mañana. Esto es otra cosa.

Dalio (uno de los inversores más exitosos de la historia) explica de forma clara:

→ Cómo funciona realmente la economía
→ Por qué las crisis se repiten una y otra vez
→ El papel que juegan la deuda y la impresión de dinero
→ Cómo piensan los grandes inversores cuando todos los demás entran en pánico

No es una charla para adivinar qué acción subirá mañana. Es algo mejor: entender el juego completo, desde la perspectiva de alguien que lleva décadas jugándolo (y ganando).

No importa si ya inviertes en bolsa o apenas estás aprendiendo sobre finanzas. Una vez que estos principios quedan claros, los mercados ya no se ven de la misma manera.

Es una de esas clases que vale la pena revisar cada cierto tiempo.

🎥 42 minutos. Gratis. Guarda este post para no perderlo.

[Min Choi @minchoi]

This Claude Code prompt is literally $300/hr senior engineer🤯

[Albia @AlbiaHossain]

One of the best designs I've seen in a while!🔥

[0xMarioNawfal @RoundtableSpace]

GEMINI 3.5 FLASH CAN DESIGN ANIMATED WEBSITES THAT LOOK LIKE THEY COST $50,000.

Someone just recorded a 25 minute tutorial showing you how to build one from scratch using Google AI Studio and Gemini 3.5 for free.

[Viktor Oddy @viktoroddy]

CLAUDE DESIGN FULL 1 HOUR COURSE

❤️‍🔥Just Recorded a Full Course on How To Build Interactive Animated Website with AI.

Bookmark this before you forget.

[0xMarioNawfal @RoundtableSpace]

AI has advanced home designs significantly in just the last few months

I’d imagine architects lives are incredibly easy now

[FHILY👑 @Oluwaphilemon1]

Сlaude Design is so cracked at marketing. 🤯

and anyone can achieve the same results.

Here's the prompt ↓

[Viktor Oddy @viktoroddy]

Claude Design is insane.

❤️‍🔥Just recorded a 18-min tutorial on how to build animated, award-winning websites with Claude Design + Opus 4.7!

[rari @0xwhrrari]

An Anthropic engineer watched me trade at a bar in SF.

Conference afterparty. Rooftop. He saw my phone - terminal was open.

"That's Polymarket."

I tell him I run 8 agents on Claude. He stops mid-sip.

"We benchmarked this internally. Claude against every wallet on the platform. Top 4%. No fine-tuning."

I ask why they never shipped it.

"Legal killed it before it left the sandbox. But the framework is public."

He opens GitHub on his phone. One repo.

https://t.co/xFkRrEjNuZ

Official agent framework. Order book parsing. Market scoring. Kelly sizing. LLM-native.

"We built our entire eval on this. Nobody outside knows."

The trick isn't win rate. It's exit timing. Claude scores wallets by disposition - how clean they close, not how often they win.

I fly home. Open Claude. One prompt.

Rebuild everything on the agents framework. Three modules. Consensus logic. Two agree - full size. One - half. Zero - skip.

Four hours later the new stack is live.

Copytrade for those who don't want to build: https://t.co/ZZrAR3mAnw

First 12 days:

+$4,700. 94 trades. 83% win rate.

Old system did $9,800 in a month. New one pacing $11,700.

$20/mo Claude. $5 VPS. That's it.

I send him a screenshot two weeks later.

"Took our team three months to get numbers like that."

Then asks me to keep his name out of it.

Done.

[gemchanger @gemchange_ltd]

https://t.co/bVArkKPyv1

[Rishabh @Rixhabh__]

Instead of watching a 2-hour movie, watch this masterclass on Claude Code from scratch:

[Travis Kavulla @TKavulla]

10 pages in and it is blowing my mind...

[The Real Estate God @TheRealEstateG6]

// Deal 1 //

You’re broke and have no skills. Everyone was here at some point. You can either whine about it or you can get moving

Reason most people think making money is hard is that they focus on how hard the end goal is (making $1MM) instead of focusing on the first step of the process – gaining the knowledge necessary to start making money

This first deal doesn’t matter at all. The only reason it matters is to “get you in the game”. Once you buy a deal you start accruing market knowledge, you start accruing relationships (lender, investor, broker, etc) – you start to become “dangerous”

You obviously want it to be a good deal but it doesn’t need to be a *great* deal. Because the purpose isn’t to make money, it’s to gain knowledge/skills/relationships so you can make real money on future deals

So what does this mean in a practical sense?

Step 1: Figure out how much equity you either have yourself or can raise

I was literally dead broke for my first deal so was only able to put $2,500 of my own money in. I reached out to my entire network and was only able to gather ~$100k. That wasn’t enough to buy a triplex all cash in my market (which is what I was targeting)

So I reached out to friends who could possibly partner up in the deal. Meant splitting the profits but didn’t matter to me. Was going to do whatever it took. If I needed to take on 10 more partners, I would have. It wasn’t about the profit, it was about the skillset

I finally found a partner and using our combined networks, we were able to muster $200k, which allowed us to buy a triplex

I’m going to assume you can raise a similar amount, but if not, no problem. Whatever amount you can raise, you back into a deal of that size. If you truly can’t raise any money, start up a real estate social media account, share your thoughts/analysis, create a network and raise the money that way

Step 2: Finding the deal

- The market: You’re going to go into a market where deals sell for less than $150k/unit. Additionally, deals in the market should be selling for 6% cap rates at minimum otherwise you’re going to find it too difficult

- You’re going to search on Zillow/Redfin for duplexes, triplexes and quads (loopnet and crexi only really work for bigger properties)

- Open up excel. Make a list of every property listed in the market. Track the price per unit it’s listed for and note the price per unit it sells for. There should be 50+ properties on this list or you’re doing it wrong. Download any and all financial information that’s included in the listing. If there’s no information in the listing, call up the broker and get it from him. By the end of this process, you should know exactly what price per unit properties sell for in the market and exactly what cap rate they trade at. You’ll be looking to buy a property below the market price per unit and above the market cap rate (stabilized yield). Simple

- Market rents: Same thing. First use the mls/costar if you have access to it. If not, get comps from brokers (still have to vet them yourself). Worst case scenario, use Zillow/apartments.com for active comps. Put them all in excel, discard any outliers and take the average. You’ll still need to run the “market rent” for each new deal you look at as each property differs slightly, but this should get you close initially

- Expenses: Use the P&L information you now have. Line each individual P&L up next to each other in excel & for each individual line item, note the cost per unit, cost per square foot & the overall expense load as a percent of revenue. Now you’ve figured out both revenue & expenses. You now should be able to create a brand new P&L for new properties you look at from scratch

- Brokers: Every time you see a broker on a listing, you’re going to call him up, ask him a few questions about the deal, tell him your criteria & ask him to add you to his mailing list. You should start receiving “passive” deal flow to your inbox every morning. Should also set Zillow/redfin alerts for your criteria. You’ll probably have to look at over 100 deals before pulling the trigger

- Underwriting. You’re going to underwrite 3-5 deals a week minimum. You’re going to underwrite them based on their stabilized yield (if you don’t know what this is, search my tweets for it). You’re looking for a 200 bps+ spread between your stabilized yield & the market cap rate. The higher the stabilized yield, the better. The goal is to understand exactly what price per unit & stabilized yield you’d buy a property in this market for so you’re ready to pull the trigger when the time comes

Step 3: Pulling the trigger

Your “buy box” should already be established, only thing left to do is find a deal that fits it. You should be tracking the aggregation sites daily, have plenty of broker contacts at this point & have a lot of deal flow hitting your inbox

When you see a deal that fits your criteria, pull the trigger

Since this is your 1st deal, you may have a hard time getting a loan. Instead of whining about this (like most people do), you can:
1. Do the deal all cash
2. Use seller financing
3. Get a guarantor to hop on the loan

Doesn’t matter which one you use, just get the deal done. As quickly as possible

Deal 1 Summary: Depending on your access to capital, your first deal should be 1-5 units in the $200k to $500k range. The timeline should be quick (doesn’t take long to renovate 5 units or less) & the profit doesn’t matter as much so I’m not even going to bother listing it (obviously if you do raise money though, be honest about the return profile up front). The idea is to “get you in the game”

[Frank ☼ Bach @zendadddy]

we will now see the gap widen between talkers and doers.

[el.cine @EHuanglu]

Wallstreet is so cooked..

this new AI agent can analyse real-time stock market data 24/7 autonomously and tell you the best time to buy or sell..

it's like having a personal finance expert in pocket

here's how it works:

[Meng To @MengTo]

I recorded a 50-min tutorial on prompting top-tier landing pages with Gemini 3

[Daniel Ch @chddaniel]

GUYS IT'S SO OVER!!! go build COMPANIES w/ claude code opus 4.5 - visually

typeform is a $1b company with +$80,000,000/yr ARR
in <1hr, I've built:
• My own Typeform
• But with Spotify's design
• It has a WORKING BACK-END
• It's got login for users (BetterAuth)
• It's got Stripe + accepts payments... I've got a SaaS
• It's got multi-translation with 1click, auto-translated by AI

This is almost a freaking company... all built by AI, in <1h

Comment "build" + RT -> randomly giving out FREE credits to build

[Ben Tossell @bentossell]

https://t.co/Ref8GgkIR5