Boston, MA
Joined April 2009
- Tweets8.1K
- Following1.1K
- Followers369
- Likes1.7K
Dominick De Loretta retweeted
SOMEONE LEAKED THE PROMPT TO CREATE TRADING AGENTS. YOU PASTE IT INTO CLAUDE CODE AND IT SETS UP THE BOT WHILE YOU GRAB A COFFEE.
No programming required. Just paste. A single prompt does it all:
Time required: 1 minute
Actions: copy and paste into Claude Code
The mistake that wipes out the accounts of 99% of traders with bots: they use the same strategy for everything.
But the market isn't the same on a quiet Tuesday as it is in the middle of a crash
1. Read the market like a quant: detect if it's calm, sideways, jittery or in freefall and switch strategies for each one
2. Handbrake: maximum drawdown of 15% and a kill switch that's checked before EVERY order
3. Test before touching the money: backtest the strategy period by period with real data and refuse to trade if it doesn't beat the market
4. Night shift: while you sleep, it analyzes what went wrong and leaves ready improvements. You decide if they're applied
5. Zero code: you paste, fill in your data and Claude handles the rest
Before, this meant a team of PhDs, mathematicians, thousands of lines of code and weeks of work.
Now it's just Ctrl+V.
Save it before they delete it.
Dominick De Loretta retweeted
Anthropic Enginner just dropped a 24-minute workshop on how to actually write prompts for Claude.
Taught by the people who built it.
Free. No registration. No paywall.
Ive seen $300 courses that dont cover what they teach in the first 8 minutes.
They break down the actual structure behind prompts that work.
Real examples. Not theory.
If you're using Claude for content, code, or automation, this is 24 minutes that will save you hours.
Watch it. Bookmark it. Reference it when youre stuck.
Dominick De Loretta retweeted
Sam Altman, CEO of OpenAI:
“You don’t need to write prompts anymore.”
In just 27 minutes, he explains how to use ChatGPT in ways most people haven’t even explored yet.
This was a talk he gave to Stanford students. A friend sent me the recording last night.
After watching it, I realized I was barely using 15% of what ChatGPT can actually do.
Watch the full talk, then check out the guide below on how to build a system that can prompt itself.
Dominick De Loretta retweeted
Grok Bot could be the first thing you hire instead of prompt
It can own a job, work inside your tools for hours and come back with finished work
In this article, I show you how
Dominick De Loretta retweeted
At 22 he said no to $1 billion from Yahoo, and opened the board meeting with one sentence:
"We're obviously not going to sell here."
His name is Mark Zuckerberg. In 2006 Facebook was two years old and still mostly for students. Yahoo offered $1 billion. His investors wanted to at least talk about it. Peter Thiel later said Zuckerberg called the meeting a formality that should take ten minutes.
Almost every early employee thought he was insane. Some of the management team left over it.
Facebook, now Meta, went on to be worth more than a trillion dollars.
Honestly, most of us would have taken it and called it the smartest move of our lives. A billion dollars at 22 is not a mistake by any normal standard.
What made the difference was that he didn't think in normal standards. He wasn't asking if $1 billion was a lot. He was asking if Facebook was worth more.
Would you have signed at 22?
Dominick De Loretta retweeted
Andrej Karpathy just dropped a 1-hour Stanford lecture on AI engineering from scratch:
"You can basically delete everything… delete everything, keep Graph"
Here are the core ideas:
10% → LLM: "I treat GPT like a general-purpose computer that can be reprogrammed at runtime"
30% → Prompt: "I give that computer a program written in natural language"
50% → Agent: "I wrap the model with a goal, context, memory, and tools that turn prediction into action"
70% → Loop: "I separate the inner loop, where the model learns from context, from the outer loop, where training updates the weights"
100% → Graph: "I organize communication as data-dependent message passing across directed graphs"
This 1-hour Stanford lecture will teach you more about AI engineering than most 100 random YouTube tutorials combined
"Just chop everything up and throw it into the mix"
Bookmark it and watch later
Then read the full article below ↓
Dominick De Loretta retweeted
Anthropic engineer:
"99% of our engineers run swarms of 300+ self-improving agents"
"Now everyone is building agentic graphs"
In a 20-minute session, an Anthropic team member breaks down how graph engineering turns isolated agents into systems that improve themselves
The real setup is Claude running through graph workflows, plan mode, and dynamic orchestration
Agents → Loops → Graphs → Self-Improving Systems
Better than most $300 graph engineering courses
Bookmark and watch the talk
Then read the article below
How about that $META MUSE ? It’s kicking buts all over the place. What can’t muse do so far?
Dominick De Loretta retweeted
Diamondback Energy $FANG: Diamondback rides oil's bounce, WTI up about 2% off Friday's close near $92 toward the mid-$94s as Iran tensions reignite a supply-risk premium. Stock had been sliding on a block sale and a Morgan Stanley downgrade despite Q2 beats and raised production guidance. @petenajarian @jonnajarian connect the dots because #ITSNOTANOPTION 📙 RebelCon 2026 Oct 19-21 Vegas 👉 tinyurl.com/8zcjumak
Dominick De Loretta retweeted
Do you understand what just got open sourced?
5,800+ quant trading strategies for FREE along with its CODEBASE
i gave these trading strategies to opus 5.5 + jev and it built me a PROFITABLE 24/7 trading agent
here is how to use the repo and build AI trading bot:
1. clone the vault and get 5,806 trading strategies (spot, perps, grid, stat arb and market making)
2. open any strategy file, the math, the parameters and the source are already inside
3. point AgenKit at that file inside Claude Code with model as opus 5.5, one command and it reads the spec as ground truth
4. it will run a full engineering team and ships a production bot with risk controls
5. Jev is the live brain, it answers buy sell or hold in 81 milliseconds, calibrated, no hallucination
6. your engine sizes the trade from that confidence and fires it before the next block
7. a hard risk layer holds every limit, nothing goes live until you approve it
8. every night Opus 5.5 reviews the fills and ships a better version, so the agent improves while you sleep
one model THINKS, the other REACTS and a strategy that sat locked in a vault now trades itself 24/7
the COMPLETE codebase and the EXACT Opus 5.5 + Jev trading system which I am running is in the article below:
Dominick De Loretta retweeted
In Q2, Sharplink doubled down in activating our treasury and ushering in a new Ethereum era.
From ETH productivity partnerships to investing in companies driving institutional adoption, this was a positive quarter of building momentum and helping accelerate the institutional adoption supercycle.
A snapshot:
→ $11.5M in quarterly revenue as of June 30
→ 888,938 ETH holdings as of August 3
→ 24,755 in cumulative staking rewards as of August 3
Dominick De Loretta retweeted
He's 16 and already an AI engineer at OpenAI making $1.8M a year - and came to Stanford to share how:
"I skipped university and made $120k a month thanks to AI agents"
00:43 - how to get an OpenAI offer at 16 with no degree
18:34 - how one agent without code brings $120k a month
32:47 - how Claude Code turns you into a $1.8M AI engineer
after watching I realized that 4 years of university gave me a degree - one agent in Claude Code would have given me $1.8M a year
save & watch - the article below is step by step how to build an agent that lands you a job at Anthropic or OpenAI.
Dominick De Loretta retweeted
Jim Simons built a $100B empire and took the math to his grave. MIT just leaked the document he never wanted published.
Two Sigma and Citadel pay analysts $500k+ a year to run exactly this framework. Independent traders were never supposed to see it.
"Practical Math for Traders." 2026 Working Note on Quantitative Trading. The internal system MIT's experimental trading lab built in secret. People close to Simons confirmed privately that Medallion Fund's early edge came from exactly this pipeline. Position sizing. Risk modeling. Portfolio construction. The complete system from raw market data to consistent returns - documented for the first time.
The path that was closed for 30 years just opened.
Here's what $500k/year quants actually do differently.
A trader with 55% win rate and 1.5 win/loss ratio. Kelly Criterion gives the optimal position size: 25% of capital per trade. Most retail traders size by feel - 50%, 80%, sometimes all-in. The math says 25%. That single number is the difference between compounding wealth and blowing up an account on a losing streak.
The document models the full trading system. Expected return. Portfolio variance. Sharpe ratio. VaR. All five calculations every desk at Goldman, Citadel, and D.E. Shaw runs before touching a position. The equity curve inside shows a trend-following strategy from 2020 to 2025. $50,000 starting capital. Drawdowns never exceeding -30%. Consistent upward curve through every crash in between.
The efficient frontier chart shows exactly where the optimal portfolio sits - the tangency point where Sharpe ratio is maximized. Every dollar of risk producing the highest possible return. That's what Medallion runs on $100B+. That's what this document teaches you to build.
Jim Simons never gave interviews about the math. He didn't need to. The document does it for him.
Watch the moment [Fig. 2 - Efficient Frontier]
Individual assets scattered. Suboptimal. Too much risk, too little return.
One point on the curve: the tangency portfolio. Maximum Sharpe ratio.
The dot that separates a $50k account that compounds from one that doesn't.
MIT Quantitative Trading Lab - Working Note, September 2026
Restricted circulation - now public
bookmark this
Dominick De Loretta retweeted
I gave him the chance to think about it and delete the tweet because I knew he wasn't mean, just a little slow
But he refused, and decided to insult me instead
Bold fucking choice.
So remember this: if Kalshi decides your behavior is embarrassing enough to fire your ass over, you chose this outcome yourself
I literally shat on polymarket < than a week ago, all you had to do what shut the fuck up
Every day I mourn the fact that @AugurProject was too early, now we’re stuck using products ran by VCs who couldn't give a shit about the product as long as growth continues
Make a market on your dogshit sites where I can bet on wether your investors would sell their moms for a bit more growth, I'll promise I'll do so much volume you won't even need to cheat
Thanks, if you enjoyed, make sure this gets the reach it deserves
I want this to serve as a fucking lesson: if you’re going to publicly talk down to people, and me out of all people, you better be certain of what you bring forward, otherwise this happens
Dominick De Loretta retweeted
sources:
CFTC filling:
02/09/2026:
cftc.gov/filings/orgrules/ru…
04/07/2026:
cftc.gov/filings/orgrules/ru…
Bloomberg article JUMP/KALSHI:
bloomberg.com/news/articles/…
Kalshi's own website for:
eth-perp, leaderboard etc:
kalshi.com/perps/kxethperp
Glossary:
alpha.kalshi.com/pro/help/gl…
Dominick De Loretta retweeted
Part 1.5
Felt bad leaving you guys with nothing tonight
So here is the undeniable proof that Kalshi is faking their PERP volume
FYI in case the data was to vanish over night, I have downloaded it and it's encrypted on 3 diff clouds
jus in case u know
Anyhow, here is an example
On ETH the exact same $5500 trade size keeps appearing over and over again
By over and over again, I mean it literally made up 48%–58% of ALL ETH PERP volume on 4 separate days
Also rumor, by rumor I mean person who told me wants to stay anonymous so I can't share proof, is that Kalshi has volume deals so MMs are forced to fuck around and do things like this
Literally incentivized to cheat
What I don't understand is why they are so bad at it kek
pro tip: if you guys want to fake volume one day, don't use the same lot size to do it over and over again
sincerely,
beni
Dominick De Loretta retweeted
Watching you pretend you didn’t see this my thread is somehow more embarrassing than the thread itself
Nothing I could say about you is as humiliating as watching you publicly demonstrate that you have absolutely fuck all to say back
But pls take your time tho
I understand, it’s hard writing around something you can’t defend
What did you have the audacity to tell me again?
Oh yeah
"Ur doing the meme buddy 😂"