Video wird geladen...

Video konnte nicht geladen werden

Zur Startseite

jane street, two sigma, man group put their actual code on github 22 repos from firms running $200 billion combined - all public, all free nobody's talking about this because nobody thought to check what's in those repos isn't just tooling - it's their mental model these firms don't...

L1vsun's profile picture

Livsun

2,740 subscribers

61,062 Aufrufe • vor 1 Monat •via X (Twitter)

Wissenschaft & TechnologieFinanzen
LIVE

Anya Rossi• Live Now

Private livecam show

0 Kommentare

Keine Kommentare verfügbar

Kommentare vom Original-Post werden hier angezeigt

Ähnliche Videos

quants don't predict markets. they stopped trying years ago retail is still asking "where is price going?" - quants switched to a different question entirely "given the current state, what's the probability distribution of outcomes" that's not a semantics game. it changes the math entirely stochastic processes model how a system evolves probabilistically - not picking a direction, mapping where each path leads and how likely it is ornstein-uhlenbeck does this for mean reversion: drift rate back to equilibrium, variance around that path same price data everybody has. different equation math is a century old, free in every applied probability textbook Bookmark and learn this before it gone you've been drawing trend lines while they ran probability distributions on same numbers it was never about data it was the question you were taught to ask
1:27

quants don't predict markets. they stopped trying years ago retail is still asking "where is price going?" - quants switched to a different question entirely "given the current state, what's the probability distribution of outcomes" that's not a semantics game. it changes the math entirely stochastic processes model how a system evolves probabilistically - not picking a direction, mapping where each path leads and how likely it is ornstein-uhlenbeck does this for mean reversion: drift rate back to equilibrium, variance around that path same price data everybody has. different equation math is a century old, free in every applied probability textbook Bookmark and learn this before it gone you've been drawing trend lines while they ran probability distributions on same numbers it was never about data it was the question you were taught to ask

Livsun

13,315 Aufrufe • vor 1 Monat

quants make $420k+/year and nobody ever wrote the guide that's not an accident people who could write it earn more by staying quiet. firms want the gap - smaller pool means more leverage on terms so here's what they actually test: probability, mental math, coding, game theory - same 4 categories, same structure, every top firm, every year. nothing about finance a kid from a state school who drilled these for 8 months has placed above ivy league finance grads at citadel - not once, routinely they don't care about your internship track or CFA they care if you can solve expected value under pressure in 30 seconds most IB analysts never even heard of the test format. they applied to the wrong door for 3 years quant track runs parallel - 8 months to be competitive for a role that pays 4-5x banking at entry information to get there is free. same problem sets recycled every year, all documented, all public Bookmark and use it for learning what's missing isn't talent nobody told you which door to knock on
1:04

quants make $420k+/year and nobody ever wrote the guide that's not an accident people who could write it earn more by staying quiet. firms want the gap - smaller pool means more leverage on terms so here's what they actually test: probability, mental math, coding, game theory - same 4 categories, same structure, every top firm, every year. nothing about finance a kid from a state school who drilled these for 8 months has placed above ivy league finance grads at citadel - not once, routinely they don't care about your internship track or CFA they care if you can solve expected value under pressure in 30 seconds most IB analysts never even heard of the test format. they applied to the wrong door for 3 years quant track runs parallel - 8 months to be competitive for a role that pays 4-5x banking at entry information to get there is free. same problem sets recycled every year, all documented, all public Bookmark and use it for learning what's missing isn't talent nobody told you which door to knock on

Livsun

24,151 Aufrufe • vor 1 Monat

the finance kids everyone envied in college made $120k at goldman the quiet math kids nobody noticed made $650k at firms most people can't name jane street. citadel. two sigma. d.e. shaw they don't recruit from linkedin. they recruit from math olympiads and competitive programming leaderboards the filter is 4 things: probability, coding, mental math, game theory same problems recycled every year, same structure every round the prep path is documented, free, and takes 8 months a kid from a state school who grinds this for 8 months has beaten ivy leaguers who showed up unprepared - credential matters less than the pattern recognition most people spend 3 years trying to break into banking for $120k the people who looked one level up spent 8 months and landed 5x the salary the information to do this has existed for years Bookmark this and upgrade your brain the gap isn't talent it's that nobody told you where to look
4:33

the finance kids everyone envied in college made $120k at goldman the quiet math kids nobody noticed made $650k at firms most people can't name jane street. citadel. two sigma. d.e. shaw they don't recruit from linkedin. they recruit from math olympiads and competitive programming leaderboards the filter is 4 things: probability, coding, mental math, game theory same problems recycled every year, same structure every round the prep path is documented, free, and takes 8 months a kid from a state school who grinds this for 8 months has beaten ivy leaguers who showed up unprepared - credential matters less than the pattern recognition most people spend 3 years trying to break into banking for $120k the people who looked one level up spent 8 months and landed 5x the salary the information to do this has existed for years Bookmark this and upgrade your brain the gap isn't talent it's that nobody told you where to look

Livsun

1,284,195 Aufrufe • vor 2 Monaten

100 billion dollars made without predicting a single price direction not a direction call, a macro bet, or a news trade in sight they asked a completely different question than every retail trader alive retail: "will this go up?" quant: "what state is the market in, and where does that state historically lead?" same data. completely different game markets aren't random - they cycle through states: trending, ranging, compressing, reversing each state has a fixed historical probability of transitioning to the next that's the whole edge. not a signal, not an indicator - just conditional probability you don't need to know what fed will do or what earnings will print you need to know which state you're in and what that state leads to 68% of the time framework has been public since 1906 Bookmark it before it gone data is free, implementation is 200 lines of python, paper costs nothing they built 100 billion off math sitting in statistics textbooks your whole life you were staring at candles while they ran probability tables
0:52

100 billion dollars made without predicting a single price direction not a direction call, a macro bet, or a news trade in sight they asked a completely different question than every retail trader alive retail: "will this go up?" quant: "what state is the market in, and where does that state historically lead?" same data. completely different game markets aren't random - they cycle through states: trending, ranging, compressing, reversing each state has a fixed historical probability of transitioning to the next that's the whole edge. not a signal, not an indicator - just conditional probability you don't need to know what fed will do or what earnings will print you need to know which state you're in and what that state leads to 68% of the time framework has been public since 1906 Bookmark it before it gone data is free, implementation is 200 lines of python, paper costs nothing they built 100 billion off math sitting in statistics textbooks your whole life you were staring at candles while they ran probability tables

Livsun

13,335 Aufrufe • vor 1 Monat

THE MOST EXPENSIVE ENGINEERING TEAMS ON EARTH JUST PUT THEIR FINANCIAL TOOLS ON GITHUB FOR FREE. Jane Street. Goldman Sachs. JP Morgan. BlackRock. Hudson River Trading. Two Sigma. D.E. Shaw. Seven firms. Seven repos. Billions in engineering talent open sourced. Save this before you scroll past it. 1. Jane Street — magic-trace 5,300 stars. Process tracer powered by Intel PT. When your profiler is blind this sees every CPU instruction. 2. Goldman Sachs — gs-quant Derivative pricing the GS traders use at their actual desks. MIT licensed. Free. 3. JP Morgan — perspective What JPMorgan traders use to watch markets in real time. A $24,000 per year terminal. Available to anyone with a GitHub account. 4. BlackRock — lcso Rust optimizer for portfolio problems. Where scipy gives up this works. Built for problems that break standard optimization libraries. 5. Hudson River Trading — corral Structured concurrency for C++20. The foundation of HFT infrastructure at one of the largest US trading firms. 6. Two Sigma — flint Time-series joins on Apache Spark with temporal tolerance. Built for billions of ticks. The data infrastructure layer behind systematic trading at scale. 7. D.E. Shaw — pyflyby Auto-import for IPython and Jupyter. D.E. Shaw also funded the development of IPython itself. The firm that built the tool is now giving you the enhancement for free. Here is what this list actually represents. These seven firms collectively employ thousands of engineers earning $300,000 to $1,000,000 per year. The tools they built to solve their hardest problems are the same tools you now have access to for free. The information asymmetry that used to separate a quant at Goldman from a developer at home just narrowed significantly. The infrastructure is free. The edge now belongs to whoever knows how to use it. Bookmark this before you pay for another financial data tool. Follow CyrilXBT for every elite engineering resource the moment it surfaces.
1:47

THE MOST EXPENSIVE ENGINEERING TEAMS ON EARTH JUST PUT THEIR FINANCIAL TOOLS ON GITHUB FOR FREE. Jane Street. Goldman Sachs. JP Morgan. BlackRock. Hudson River Trading. Two Sigma. D.E. Shaw. Seven firms. Seven repos. Billions in engineering talent open sourced. Save this before you scroll past it. 1. Jane Street — magic-trace 5,300 stars. Process tracer powered by Intel PT. When your profiler is blind this sees every CPU instruction. 2. Goldman Sachs — gs-quant Derivative pricing the GS traders use at their actual desks. MIT licensed. Free. 3. JP Morgan — perspective What JPMorgan traders use to watch markets in real time. A $24,000 per year terminal. Available to anyone with a GitHub account. 4. BlackRock — lcso Rust optimizer for portfolio problems. Where scipy gives up this works. Built for problems that break standard optimization libraries. 5. Hudson River Trading — corral Structured concurrency for C++20. The foundation of HFT infrastructure at one of the largest US trading firms. 6. Two Sigma — flint Time-series joins on Apache Spark with temporal tolerance. Built for billions of ticks. The data infrastructure layer behind systematic trading at scale. 7. D.E. Shaw — pyflyby Auto-import for IPython and Jupyter. D.E. Shaw also funded the development of IPython itself. The firm that built the tool is now giving you the enhancement for free. Here is what this list actually represents. These seven firms collectively employ thousands of engineers earning $300,000 to $1,000,000 per year. The tools they built to solve their hardest problems are the same tools you now have access to for free. The information asymmetry that used to separate a quant at Goldman from a developer at home just narrowed significantly. The infrastructure is free. The edge now belongs to whoever knows how to use it. Bookmark this before you pay for another financial data tool. Follow CyrilXBT for every elite engineering resource the moment it surfaces.

CyrilXBT

42,296 Aufrufe • vor 2 Monaten

a citadel options trader told me the one concept they test first in every quant interview and it's been sitting on a free website for years not a hedge fund textbook, not a $3,000 prep program. a free course syllabus - options greeks, volatility, quizzes - publicly available, almost nobody applying has ever opened it concept is expected value across a probability distribution retail looks at a chart and asks which direction. quant looks at expected payout across every possible outcome and asks if that number beats the cost of the trade - completely different question options pricing is just EV made rigorous fair value of any position = sum of (each outcome's probability x its payoff), discounted back. that formula is in every intro stats course and every free options curriculum these firms post publicly citadel's first round isn't a stock pitch or a DCF it's a market-making problem: "set me a bid and ask on a coin flip" if you can solve that fast and size it correctly, you can price any derivative on earth prep is documented in 6 categories: probability, greeks, volatility, mental math, coding, microstructure firms don't want you pattern-matching to old trades. they want raw EV instinct - and that's in free courses that have been online for years entry-level quant traders at these firms start at $300k. senior traders clear $650k+ most people never make it past round 1. not because they weren't smart - because nobody told them what the test was actually measuring Bookmark this they kept you reading charts while they were drilling expected value at 2am
0:56

a citadel options trader told me the one concept they test first in every quant interview and it's been sitting on a free website for years not a hedge fund textbook, not a $3,000 prep program. a free course syllabus - options greeks, volatility, quizzes - publicly available, almost nobody applying has ever opened it concept is expected value across a probability distribution retail looks at a chart and asks which direction. quant looks at expected payout across every possible outcome and asks if that number beats the cost of the trade - completely different question options pricing is just EV made rigorous fair value of any position = sum of (each outcome's probability x its payoff), discounted back. that formula is in every intro stats course and every free options curriculum these firms post publicly citadel's first round isn't a stock pitch or a DCF it's a market-making problem: "set me a bid and ask on a coin flip" if you can solve that fast and size it correctly, you can price any derivative on earth prep is documented in 6 categories: probability, greeks, volatility, mental math, coding, microstructure firms don't want you pattern-matching to old trades. they want raw EV instinct - and that's in free courses that have been online for years entry-level quant traders at these firms start at $300k. senior traders clear $650k+ most people never make it past round 1. not because they weren't smart - because nobody told them what the test was actually measuring Bookmark this they kept you reading charts while they were drilling expected value at 2am

Livsun

25,988 Aufrufe • vor 2 Monaten

hedge funds extract their edge before the candle closes not from price, from what's moving behind it signal is called order flow toxicity every large informed trader disguises their order as retail noise - 40-80 small orders, timed to look random to a human, invisible. to an LSTM trained on cancel-to-fill ratios - readable in 400 milliseconds two sigma doesn't ask "will price go up?" - they ask "is someone with information buying right now, before the market reprices?" detection model is 3 layers: > input: bid-ask spread velocity, cancel-to-fill ratio, trade size entropy > hidden: LSTM over 500ms rolling window > output: toxicity score 0-1 above 0.67: informed flow detected. enter with it below 0.67: noise. skip or fade 74% directional accuracy. 11-minute avg hold. runs on any level 2 feed data is free - CBOE publishes it. architecture is in every ML textbook since 2012 they've been running this since you were drawing trendlines Bookmark it and add in your strategy bottleneck was never the data - it was the question you were asking
0:48

hedge funds extract their edge before the candle closes not from price, from what's moving behind it signal is called order flow toxicity every large informed trader disguises their order as retail noise - 40-80 small orders, timed to look random to a human, invisible. to an LSTM trained on cancel-to-fill ratios - readable in 400 milliseconds two sigma doesn't ask "will price go up?" - they ask "is someone with information buying right now, before the market reprices?" detection model is 3 layers: > input: bid-ask spread velocity, cancel-to-fill ratio, trade size entropy > hidden: LSTM over 500ms rolling window > output: toxicity score 0-1 above 0.67: informed flow detected. enter with it below 0.67: noise. skip or fade 74% directional accuracy. 11-minute avg hold. runs on any level 2 feed data is free - CBOE publishes it. architecture is in every ML textbook since 2012 they've been running this since you were drawing trendlines Bookmark it and add in your strategy bottleneck was never the data - it was the question you were asking

Livsun

11,425 Aufrufe • vor 2 Monaten

every time you hit buy, the market maker on the other side already solved for two numbers not on the asset - on you specifically the first: their internal fair value. not the midpoint you see on screen their number is adjusted for the last 30 seconds of order flow, their current inventory imbalance and which volatility regime the model thinks the market is in it's different from the displayed price, and they never show it to you the second: your adverse selection probability. a live estimate of whether you're informed or just emotional if the model flags you as a potential informed trader, the spread widens before your fill - automatically, invisibly, before you notice anything changed that bad fill you blamed on "low liquidity"? sometimes it's a market maker's algorithm deciding you look too dangerous to fill at the quoted price this is Glosten-Milgrom - published 1985, sits in every market microstructure textbook on earth, runs inside every serious market making operation that's ever quoted you a price Bookmark this feature before it lost when you click buy, you're answering a question you didn't know was being asked retail spends all its time thinking about where price is going the desk filling your order is spending that same time thinking about you
0:30

every time you hit buy, the market maker on the other side already solved for two numbers not on the asset - on you specifically the first: their internal fair value. not the midpoint you see on screen their number is adjusted for the last 30 seconds of order flow, their current inventory imbalance and which volatility regime the model thinks the market is in it's different from the displayed price, and they never show it to you the second: your adverse selection probability. a live estimate of whether you're informed or just emotional if the model flags you as a potential informed trader, the spread widens before your fill - automatically, invisibly, before you notice anything changed that bad fill you blamed on "low liquidity"? sometimes it's a market maker's algorithm deciding you look too dangerous to fill at the quoted price this is Glosten-Milgrom - published 1985, sits in every market microstructure textbook on earth, runs inside every serious market making operation that's ever quoted you a price Bookmark this feature before it lost when you click buy, you're answering a question you didn't know was being asked retail spends all its time thinking about where price is going the desk filling your order is spending that same time thinking about you

Livsun

18,906 Aufrufe • vor 1 Monat

every day before market open, the CBOE publishes exactly where market makers are forced to buy and sell not a theory, not a pattern - a number that actually moves price it's called dealer gamma exposure - GEX - and it creates real price magnets quants have traded against for years here's why it works when you buy a call option, a market maker sells it to you and has to hedge that hedge shifts as price moves - they buy when it rises, sell when it falls negative GEX zones are where they sell into every rally, cap every breakout, crush every momentum move strikes with highest open interest aren't random - they're gravitational markets don't break through those levels cleanly because a billion-dollar dealer is actively hedging against it quants call them gamma walls data is free, published every morning on the CBOE site - no API, no Bloomberg terminal retail calls those same levels "resistance" and chalks it up to vibes bookmark this before it lost it's not vibes. it's math someone is paid $400k/year to understand while you do it by eyeball the whole time you thought technical analysis was detecting patterns you were just watching the shadow of dealer hedging and calling it insight
1:10

every day before market open, the CBOE publishes exactly where market makers are forced to buy and sell not a theory, not a pattern - a number that actually moves price it's called dealer gamma exposure - GEX - and it creates real price magnets quants have traded against for years here's why it works when you buy a call option, a market maker sells it to you and has to hedge that hedge shifts as price moves - they buy when it rises, sell when it falls negative GEX zones are where they sell into every rally, cap every breakout, crush every momentum move strikes with highest open interest aren't random - they're gravitational markets don't break through those levels cleanly because a billion-dollar dealer is actively hedging against it quants call them gamma walls data is free, published every morning on the CBOE site - no API, no Bloomberg terminal retail calls those same levels "resistance" and chalks it up to vibes bookmark this before it lost it's not vibes. it's math someone is paid $400k/year to understand while you do it by eyeball the whole time you thought technical analysis was detecting patterns you were just watching the shadow of dealer hedging and calling it insight

Livsun

69,577 Aufrufe • vor 1 Monat

OpenAI just created a $10 billion company whose ONLY job is forcing businesses to use AI. And they're literally guaranteeing investors a 17.5% annual return to make it happen. It's called "The Deployment Company." OpenAI finalized it yesterday with 19 investors including TPG, SoftBank, Bain Capital, Brookfield, and Advent International. Here's the structure: OpenAI puts in $1.5 billion. The private equity firms put in $4 billion. In exchange, those PE firms open up their 2,000+ portfolio companies as a CAPTIVE customer base for OpenAI's products. OpenAI then embeds teams of engineers directly inside those companies, Palantir-style, to integrate their tools into daily operations. And here's the big red flag in all of this: OpenAI is GUARANTEEING those PE firms a 17.5% annual return over five years. That means even if the companies in the portfolio don't want AI, don't need AI, or get zero value from AI, OpenAI is still on the hook to pay those returns. Think about what that means for a second. OpenAI is so desperate for enterprise adoption that they're paying Wall Street to force their product into thousands of businesses. They've essentially turned private equity firms into a distribution cartel with a guaranteed commission. This has NEVER been done before in enterprise software. No software company in history has guaranteed above-market returns to financial sponsors just to get their product installed. And it gets crazier: Within MINUTES of OpenAI's announcement, Anthropic announced their own version. A $1.5 billion joint venture with Blackstone, Goldman Sachs, and Hellman & Friedman. Same playbook. Two companies worth a combined $1+ TRILLION in private valuation both concluded on the same day that organic demand for their products is not growing fast enough. If enterprises were lining up to buy AI on their own, you wouldn't need to bribe private equity firms with guaranteed returns to shove it into their portfolios. You would just sell it normally like every other software company in history. But they can't. Because the gap between what AI companies PROMISE and what enterprises actually experience is still enormous. OpenAI's COO Brad Lightcap just moved into a new role specifically to lead this push. They've also signed "Frontier Alliances" with major consulting firms to embed AI through professional services channels. Every move they're making screams the same thing: We have a demand problem. And this is all happening right before OpenAI tries to IPO at $850 billion. If they can show Wall Street that 2,000+ companies are "using OpenAI products" through this PE distribution channel, it inflates their enterprise metrics right before the roadshow. Doesn't matter if those companies actually need it or if it creates real value. What matters is the number on the S-1. This is the AI playbook entering its most dangerous phase. The tech is real but the business model is being held together by financial engineering, guaranteed returns, and captive distribution deals that look more like a pharmaceutical company paying doctors to prescribe their drug than a software company earning customers on merit. And both OpenAI and Anthropic admitted it on the same day.
0:54

OpenAI just created a $10 billion company whose ONLY job is forcing businesses to use AI. And they're literally guaranteeing investors a 17.5% annual return to make it happen. It's called "The Deployment Company." OpenAI finalized it yesterday with 19 investors including TPG, SoftBank, Bain Capital, Brookfield, and Advent International. Here's the structure: OpenAI puts in $1.5 billion. The private equity firms put in $4 billion. In exchange, those PE firms open up their 2,000+ portfolio companies as a CAPTIVE customer base for OpenAI's products. OpenAI then embeds teams of engineers directly inside those companies, Palantir-style, to integrate their tools into daily operations. And here's the big red flag in all of this: OpenAI is GUARANTEEING those PE firms a 17.5% annual return over five years. That means even if the companies in the portfolio don't want AI, don't need AI, or get zero value from AI, OpenAI is still on the hook to pay those returns. Think about what that means for a second. OpenAI is so desperate for enterprise adoption that they're paying Wall Street to force their product into thousands of businesses. They've essentially turned private equity firms into a distribution cartel with a guaranteed commission. This has NEVER been done before in enterprise software. No software company in history has guaranteed above-market returns to financial sponsors just to get their product installed. And it gets crazier: Within MINUTES of OpenAI's announcement, Anthropic announced their own version. A $1.5 billion joint venture with Blackstone, Goldman Sachs, and Hellman & Friedman. Same playbook. Two companies worth a combined $1+ TRILLION in private valuation both concluded on the same day that organic demand for their products is not growing fast enough. If enterprises were lining up to buy AI on their own, you wouldn't need to bribe private equity firms with guaranteed returns to shove it into their portfolios. You would just sell it normally like every other software company in history. But they can't. Because the gap between what AI companies PROMISE and what enterprises actually experience is still enormous. OpenAI's COO Brad Lightcap just moved into a new role specifically to lead this push. They've also signed "Frontier Alliances" with major consulting firms to embed AI through professional services channels. Every move they're making screams the same thing: We have a demand problem. And this is all happening right before OpenAI tries to IPO at $850 billion. If they can show Wall Street that 2,000+ companies are "using OpenAI products" through this PE distribution channel, it inflates their enterprise metrics right before the roadshow. Doesn't matter if those companies actually need it or if it creates real value. What matters is the number on the S-1. This is the AI playbook entering its most dangerous phase. The tech is real but the business model is being held together by financial engineering, guaranteed returns, and captive distribution deals that look more like a pharmaceutical company paying doctors to prescribe their drug than a software company earning customers on merit. And both OpenAI and Anthropic admitted it on the same day.

Ricardo

52,664 Aufrufe • vor 3 Monaten

whoever leak this have titanium balls at 4am, while you sleep, a quant fund has re-weighted 300 signals, identified names to fade at open, and priced in exactly what you're about to do next number buried in here: they model retail order flow 40 minutes before the bell with 73% directional accuracy - not from secret data, from public futures positioning that's not prediction. that's your loss, booked before you wake up here's what the article unpacks about how the 4am pipeline actually runs - and where your money goes: - factor signals get re-ranked nightly on vol-adjusted returns from the prior session and overnight futures - by 9:30 every quant desk has a fresh edge score while you're reading yesterday's close - "gap up, buy the open" isn't alpha: it's one of most modeled retail behaviors in existence, and quant desk has been long since 2am and is selling into your confidence at 9:31 - they don't read same earnings release you do at 8am - automated parsers ran it 5 hours earlier, positions were set, name is already priced before it hits your news feed - order flow imbalance from asian markets predicts US open direction - quant desks weight it explicitly, most retail traders have never heard of it, and both groups see same public exchange data - vol regime they set at 4am changes position sizing by 3x - high-volatility environment means they hold a third the size and harvest from tighter edges; you're just guessing at size - news sentiment parsers score every overnight headline on a factor model trained on how similar stories moved same names in the past - by the time CNBC covers it, they've already positioned - they don't ask "which way?" at open. they ask "which way will retail push this, and where does it exhaust?" - that question alone is worth more than any indicator you run catch: not one input above is proprietary - futures prices are free, exchange data is public, and the academic papers on retail order flow sat on government websites since 2003 they read it, built on top of it, and you were never told where to look every time you log in at 9:29 feeling sharp, that session started 5 hours ago without you read what 4am actually looks like ↓
0:30

whoever leak this have titanium balls at 4am, while you sleep, a quant fund has re-weighted 300 signals, identified names to fade at open, and priced in exactly what you're about to do next number buried in here: they model retail order flow 40 minutes before the bell with 73% directional accuracy - not from secret data, from public futures positioning that's not prediction. that's your loss, booked before you wake up here's what the article unpacks about how the 4am pipeline actually runs - and where your money goes: - factor signals get re-ranked nightly on vol-adjusted returns from the prior session and overnight futures - by 9:30 every quant desk has a fresh edge score while you're reading yesterday's close - "gap up, buy the open" isn't alpha: it's one of most modeled retail behaviors in existence, and quant desk has been long since 2am and is selling into your confidence at 9:31 - they don't read same earnings release you do at 8am - automated parsers ran it 5 hours earlier, positions were set, name is already priced before it hits your news feed - order flow imbalance from asian markets predicts US open direction - quant desks weight it explicitly, most retail traders have never heard of it, and both groups see same public exchange data - vol regime they set at 4am changes position sizing by 3x - high-volatility environment means they hold a third the size and harvest from tighter edges; you're just guessing at size - news sentiment parsers score every overnight headline on a factor model trained on how similar stories moved same names in the past - by the time CNBC covers it, they've already positioned - they don't ask "which way?" at open. they ask "which way will retail push this, and where does it exhaust?" - that question alone is worth more than any indicator you run catch: not one input above is proprietary - futures prices are free, exchange data is public, and the academic papers on retail order flow sat on government websites since 2003 they read it, built on top of it, and you were never told where to look every time you log in at 9:29 feeling sharp, that session started 5 hours ago without you read what 4am actually looks like ↓

Livsun

12,857 Aufrufe • vor 7 Tagen

a quant who started as a technical trader explained what changed everything for him he stopped trying to predict where price goes he started measuring the probability of each market state and betting only when the math was asymmetric that shift took him from drawing trendlines to writing models at 22 the realization was simple but brutal: technical analysis gives you a narrative quantitative analysis gives you a number one feels right the other is testable, repeatable, and either works or doesn't across 10,000 trades the setup on his desk tells the whole story no TradingView. no candlestick charts. just code, data, and a terminal he didn't go to MIT he didn't intern at Goldman he learned Python, statistics, and probability theory on his own and built something that actually worked > the math: free in any stats textbook > the data: free on Yahoo Finance, FRED, exchange APIs > the code: Python, 200 lines, running on a laptop > the barrier: not intelligence. just knowing this path exists most people spend years staring at indicators from the 1970s wondering why they can't find consistency the answer was never a better indicator it was a completely different framework one that treats trading as a math problem, not a prediction game full breakdown in the video below
4:36

a quant who started as a technical trader explained what changed everything for him he stopped trying to predict where price goes he started measuring the probability of each market state and betting only when the math was asymmetric that shift took him from drawing trendlines to writing models at 22 the realization was simple but brutal: technical analysis gives you a narrative quantitative analysis gives you a number one feels right the other is testable, repeatable, and either works or doesn't across 10,000 trades the setup on his desk tells the whole story no TradingView. no candlestick charts. just code, data, and a terminal he didn't go to MIT he didn't intern at Goldman he learned Python, statistics, and probability theory on his own and built something that actually worked > the math: free in any stats textbook > the data: free on Yahoo Finance, FRED, exchange APIs > the code: Python, 200 lines, running on a laptop > the barrier: not intelligence. just knowing this path exists most people spend years staring at indicators from the 1970s wondering why they can't find consistency the answer was never a better indicator it was a completely different framework one that treats trading as a math problem, not a prediction game full breakdown in the video below

delost

112,271 Aufrufe • vor 1 Monat

retail has 10,000 indicators and still loses 80% of the time quants have one number they pull every morning before touching anything else it's called options skew - the gap between how much downside protection costs versus upside exposure on any liquid name CBOE publishes it free, every 15 minutes, on every major ticker since 2004 when put vol runs 8+ points above call vol on SPY, institutional desks are paying premium to hedge hard - that's not noise, that's where money thinks risk actually is backtested 2009 to 2024: when skew spikes 1.5 sigma above its 90-day baseline, SPY underperforms next 15 sessions 68% of the time not predicting direction - reading where billion-dollar desks are quietly paying to protect themselves size with kelly at 68% edge, enter short, exit in three weeks retail spent years memorizing moving average crossovers quants spent the same years reading a free number nobody told them existed same market. completely different question bookmark this data was public the whole time, you just didn't know to look for it
0:30

retail has 10,000 indicators and still loses 80% of the time quants have one number they pull every morning before touching anything else it's called options skew - the gap between how much downside protection costs versus upside exposure on any liquid name CBOE publishes it free, every 15 minutes, on every major ticker since 2004 when put vol runs 8+ points above call vol on SPY, institutional desks are paying premium to hedge hard - that's not noise, that's where money thinks risk actually is backtested 2009 to 2024: when skew spikes 1.5 sigma above its 90-day baseline, SPY underperforms next 15 sessions 68% of the time not predicting direction - reading where billion-dollar desks are quietly paying to protect themselves size with kelly at 68% edge, enter short, exit in three weeks retail spent years memorizing moving average crossovers quants spent the same years reading a free number nobody told them existed same market. completely different question bookmark this data was public the whole time, you just didn't know to look for it

Livsun

77,040 Aufrufe • vor 4 Tagen

hedge funds spent $2M to build what you could replicate in a weekend they just knew it was possible first - that was their only advantage a researcher at d.e. shaw handed me a paper in 2023 two years of quant labor, internal "discovery" - original source was an MIT lecture from 1998, public, free, searchable mechanism is cointegration two assets drift independently, but their spread behaves like a rubber band - stationary, mean-reverting, measurable retail asks: will this go up quant asks: is this spread 2.3 standard deviations outside its 5-year mean - and what's historical reversion rate from here completely different question. second one has an answer run it on GLD/GDX, SPY/QQQ, XLE/XOM - data goes back 20 years, free on yahoo finance spread reverts within 8 trading days 71% of the time when it crosses 2 sigma size with Kelly on that 71%, take the trade when math clears, exit at mean paper they spent two years on is engle & granger, 1987 - won a nobel prize Bookmark before it deleted it's in every econometrics textbook on earth, chapter 4 they didn't hide the math from you they just never told you it existed
1:02

hedge funds spent $2M to build what you could replicate in a weekend they just knew it was possible first - that was their only advantage a researcher at d.e. shaw handed me a paper in 2023 two years of quant labor, internal "discovery" - original source was an MIT lecture from 1998, public, free, searchable mechanism is cointegration two assets drift independently, but their spread behaves like a rubber band - stationary, mean-reverting, measurable retail asks: will this go up quant asks: is this spread 2.3 standard deviations outside its 5-year mean - and what's historical reversion rate from here completely different question. second one has an answer run it on GLD/GDX, SPY/QQQ, XLE/XOM - data goes back 20 years, free on yahoo finance spread reverts within 8 trading days 71% of the time when it crosses 2 sigma size with Kelly on that 71%, take the trade when math clears, exit at mean paper they spent two years on is engle & granger, 1987 - won a nobel prize Bookmark before it deleted it's in every econometrics textbook on earth, chapter 4 they didn't hide the math from you they just never told you it existed

Livsun

26,816 Aufrufe • vor 2 Monaten

a quant at a prop firm showed me a 5x5 grid on a napkin said: > this is our entire edge. we don't predict price. we predict which box the market is in and where that box historically leads i didn't understand it for weeks. then it clicked never looked at a chart the same way since grid is called a Markov Chain transition matrix. the math is from 1906, it's in every probability textbook on earth and hedge funds use it because it asks a completely different question than retail traders ever ask retail: will this go up or down quant: what state is this market in, and where does this state typically go every market lives in one of maybe 5-6 states at any given moment tight range, volatility compression, trending with momentum, post-spike reversal, pre-breakout coil not random labels - clusters you identify from actual data using volatility, volume, and momentum readings stacked together once you have the states, you build the matrix: P(state 2 -> state 4) = 73% P(state 4 -> state 1) = 61% P(state 1 -> state 3) = 68% each cell is a historical probability. now when the market is in state 2, you're not guessing you're betting on 73% historical completion. you size it with Kelly. you take the trade when the math says to, not when it feels right i built this on BTC using 2 years of 4-hour data. identified 5 states one i labeled "volatility compression below 20-day mean for 6+ consecutive candles" transitioned to a directional move above 1.8 ATR in 71% of cases average reward/risk on those trades: 5.4 that's not prediction. that's reading a probability table the market keeps filling in for you every single day the part that should bother you: the data to build this is free. the framework is in any quant textbook python to implement it is maybe 200 lines what Renaissance Technologies has that you don't isn't secret data or proprietary signals it's this framework applied to higher-resolution data with more sophisticated state definitions you're not missing information you're asking the wrong question every single time you open a chart
0:26

a quant at a prop firm showed me a 5x5 grid on a napkin said: > this is our entire edge. we don't predict price. we predict which box the market is in and where that box historically leads i didn't understand it for weeks. then it clicked never looked at a chart the same way since grid is called a Markov Chain transition matrix. the math is from 1906, it's in every probability textbook on earth and hedge funds use it because it asks a completely different question than retail traders ever ask retail: will this go up or down quant: what state is this market in, and where does this state typically go every market lives in one of maybe 5-6 states at any given moment tight range, volatility compression, trending with momentum, post-spike reversal, pre-breakout coil not random labels - clusters you identify from actual data using volatility, volume, and momentum readings stacked together once you have the states, you build the matrix: P(state 2 -> state 4) = 73% P(state 4 -> state 1) = 61% P(state 1 -> state 3) = 68% each cell is a historical probability. now when the market is in state 2, you're not guessing you're betting on 73% historical completion. you size it with Kelly. you take the trade when the math says to, not when it feels right i built this on BTC using 2 years of 4-hour data. identified 5 states one i labeled "volatility compression below 20-day mean for 6+ consecutive candles" transitioned to a directional move above 1.8 ATR in 71% of cases average reward/risk on those trades: 5.4 that's not prediction. that's reading a probability table the market keeps filling in for you every single day the part that should bother you: the data to build this is free. the framework is in any quant textbook python to implement it is maybe 200 lines what Renaissance Technologies has that you don't isn't secret data or proprietary signals it's this framework applied to higher-resolution data with more sophisticated state definitions you're not missing information you're asking the wrong question every single time you open a chart

Livsun

188,258 Aufrufe • vor 2 Monaten

John Fredericks explains the fraud in Minnesota is not going to stop because Democrats are involved and it’s a trade off for the entire Somalian community’s votes “We've got a whistleblower that is a TSA agent at the Minneapolis International Airport who said on a weekly basis, two Somali men would go through with bags and bags of cash, hundreds and hundreds of millions of dollars, go right in the baggage, put it on the airplane, take it out of the country. Nobody said anything, even though they had to log it in. All the cops looked the other way. Everybody looked the other way. Why? Because they're all in the take. There's no doubt about it. You've got videos now with this young journalist out there. You've got videos that he put up where basically Somalian parents go in with their children, check them in, get an envelope for cash and leave, right? So they just keep stealing the money in perpetuity and why not? Now, here's why it's not going to stop — T is why the Democrats are going to continue to defend them. You have to understand the way the Democrats look at this, not like we do. These are their future voters. They need illegals in here voting. This is the only way they stay in power. So this is their voting block. This is why they're resisting all the immigration deportations, because every time somebody is deported, they're just losing one of their voters and one of their foot soldiers. So that's what the problem is. So they're gonna look the other way, the Florida's gonna go on, they're gonna keep stealing your money every day, no one cares, and Minnesota politicians are gonna protect them because they elect them.“
2:58

John Fredericks explains the fraud in Minnesota is not going to stop because Democrats are involved and it’s a trade off for the entire Somalian community’s votes “We've got a whistleblower that is a TSA agent at the Minneapolis International Airport who said on a weekly basis, two Somali men would go through with bags and bags of cash, hundreds and hundreds of millions of dollars, go right in the baggage, put it on the airplane, take it out of the country. Nobody said anything, even though they had to log it in. All the cops looked the other way. Everybody looked the other way. Why? Because they're all in the take. There's no doubt about it. You've got videos now with this young journalist out there. You've got videos that he put up where basically Somalian parents go in with their children, check them in, get an envelope for cash and leave, right? So they just keep stealing the money in perpetuity and why not? Now, here's why it's not going to stop — T is why the Democrats are going to continue to defend them. You have to understand the way the Democrats look at this, not like we do. These are their future voters. They need illegals in here voting. This is the only way they stay in power. So this is their voting block. This is why they're resisting all the immigration deportations, because every time somebody is deported, they're just losing one of their voters and one of their foot soldiers. So that's what the problem is. So they're gonna look the other way, the Florida's gonna go on, they're gonna keep stealing your money every day, no one cares, and Minnesota politicians are gonna protect them because they elect them.“

Wall Street Apes

54,008 Aufrufe • vor 7 Monaten

THIS REPO IS F*CKING GOLD i genuinely wasn't planning to share this last night i stumbled onto a github repo that feels like someone accidentally left a quant firm's bookshelf unlocked hundreds of resources, papers people usually find years later market microstructure, stat arb, execution models, order flow books, datasets, lectures, research, code... all in one place what pissed me off wasn't even repo itself it was realizing how many people are still grinding through recycled twitter threads and $500 courses while this stuff has been sitting in public for years that's how this game works most people don't lose because they're lazy they lose because they never even see same information people at top quietly build on by time everyone starts talking about something, edge is already gone meanwhile somebody downloads one repo, spends 3 months reading instead of doomscrolling, and ends up seeing market completely differently hardest part isn't learning Bookmark before it losted forever it's finding what actually matters before everyone else does
1:01

THIS REPO IS F*CKING GOLD i genuinely wasn't planning to share this last night i stumbled onto a github repo that feels like someone accidentally left a quant firm's bookshelf unlocked hundreds of resources, papers people usually find years later market microstructure, stat arb, execution models, order flow books, datasets, lectures, research, code... all in one place what pissed me off wasn't even repo itself it was realizing how many people are still grinding through recycled twitter threads and $500 courses while this stuff has been sitting in public for years that's how this game works most people don't lose because they're lazy they lose because they never even see same information people at top quietly build on by time everyone starts talking about something, edge is already gone meanwhile somebody downloads one repo, spends 3 months reading instead of doomscrolling, and ends up seeing market completely differently hardest part isn't learning Bookmark before it losted forever it's finding what actually matters before everyone else does

Livsun

26,709 Aufrufe • vor 10 Tagen

a prop trader from chicago made $847k in 180 days just by asking one question every single morning that 99% of traders never ask he didn't build a new model didn't touch machine learning just opened excel and spent 8 minutes on one calculationthe question: what state is the market in right now, and where does it statistically go next most traders ask "will this go up or down". that's 50/50 he started asking "is the market trending, ranging, or reversing" and then looked at the historical probability of each transitionturns out markets don't flip randomly they cycle through states. each state has a fixed probability of shifting to the next onehe built a 5x5 grid on a napkin: trending up -> 68% stays trending, 21% flips to range, 11% reverses ranging -> 54% stays range, 28% breaks up, 18% breaks down trending down -> 61% stays falling, 24% to range, 15% reverses he didn't predict direction he just calculated which state had the highest expected value and sized the position with kelly criterion that's the entire edgethe framework is from 1906 - andrei markov. free in every probability textbook on earthrenaissance technologies has been running this since 1988 37 years of 66% annual returnsdata costs nothing - yahoo finance, federal reserve, any broker implementation is 200 lines of python what separates him from the retail traders losing money isn't intelligence or capital or luck it's that he was willing to think differently about the same data everyone else sees every single day they kept you staring at candles while the people who got it were reading transition matrices bookmark this - you're either asking the wrong question or you're not asking it at all
0:58

a prop trader from chicago made $847k in 180 days just by asking one question every single morning that 99% of traders never ask he didn't build a new model didn't touch machine learning just opened excel and spent 8 minutes on one calculationthe question: what state is the market in right now, and where does it statistically go next most traders ask "will this go up or down". that's 50/50 he started asking "is the market trending, ranging, or reversing" and then looked at the historical probability of each transitionturns out markets don't flip randomly they cycle through states. each state has a fixed probability of shifting to the next onehe built a 5x5 grid on a napkin: trending up -> 68% stays trending, 21% flips to range, 11% reverses ranging -> 54% stays range, 28% breaks up, 18% breaks down trending down -> 61% stays falling, 24% to range, 15% reverses he didn't predict direction he just calculated which state had the highest expected value and sized the position with kelly criterion that's the entire edgethe framework is from 1906 - andrei markov. free in every probability textbook on earthrenaissance technologies has been running this since 1988 37 years of 66% annual returnsdata costs nothing - yahoo finance, federal reserve, any broker implementation is 200 lines of python what separates him from the retail traders losing money isn't intelligence or capital or luck it's that he was willing to think differently about the same data everyone else sees every single day they kept you staring at candles while the people who got it were reading transition matrices bookmark this - you're either asking the wrong question or you're not asking it at all

Livsun

41,634 Aufrufe • vor 1 Monat

NOBODY wants to send their data to Google or OpenAI. Yet here we are, shipping proprietary code, customer information, and sensitive business logic to closed-source APIs we don't control. While everyone's chasing the latest closed-source releases, open-source models are quietly becoming the practical choice for many production systems. Here's what everyone is missing: Open-source models are catching up fast, and they bring something the big labs can't: privacy, speed, and control. I built a playground to test this myself. Used CometML's Opik to evaluate models on real code generation tasks - testing correctness, readability, and best practices against actual GitHub repos. Here's what surprised me: OSS models like MiniMax-M2, Kimi k2 performed on par with the likes of Gemini 3 and Claude Sonnet 4.5 on most tasks. But practically MiniMax-M2 turns out to be a winner as it's twice as fast and 12x cheaper when you compare it to models like Sonnet 4.5. Well, this isn't just about saving money. When your model is smaller and faster, you can deploy it in places closed-source APIs can't reach: ↳ Real-time applications that need sub-second responses ↳ Edge devices where latency kills user experience ↳ On-premise systems where data never leaves your infrastructure MiniMax-M2 runs with only 10B activated parameters. That efficiency means lower latency, higher throughput, and the ability to handle interactive agents without breaking the bank. The intelligence-to-cost ratio here changes what's possible. You're not choosing between quality and affordability anymore. You're not sacrificing privacy for performance. The gap is closing, and in many cases, it's already closed. If you're building anything that needs to be fast, private, or deployed at scale, it's worth taking a look at what's now available. MiniMax-M2 is 100% open-source, free for developers right now. I have shared the link to their GitHub repo in the next tweet. You will also find the code for the playground and evaluations I've done.
0:48

NOBODY wants to send their data to Google or OpenAI. Yet here we are, shipping proprietary code, customer information, and sensitive business logic to closed-source APIs we don't control. While everyone's chasing the latest closed-source releases, open-source models are quietly becoming the practical choice for many production systems. Here's what everyone is missing: Open-source models are catching up fast, and they bring something the big labs can't: privacy, speed, and control. I built a playground to test this myself. Used CometML's Opik to evaluate models on real code generation tasks - testing correctness, readability, and best practices against actual GitHub repos. Here's what surprised me: OSS models like MiniMax-M2, Kimi k2 performed on par with the likes of Gemini 3 and Claude Sonnet 4.5 on most tasks. But practically MiniMax-M2 turns out to be a winner as it's twice as fast and 12x cheaper when you compare it to models like Sonnet 4.5. Well, this isn't just about saving money. When your model is smaller and faster, you can deploy it in places closed-source APIs can't reach: ↳ Real-time applications that need sub-second responses ↳ Edge devices where latency kills user experience ↳ On-premise systems where data never leaves your infrastructure MiniMax-M2 runs with only 10B activated parameters. That efficiency means lower latency, higher throughput, and the ability to handle interactive agents without breaking the bank. The intelligence-to-cost ratio here changes what's possible. You're not choosing between quality and affordability anymore. You're not sacrificing privacy for performance. The gap is closing, and in many cases, it's already closed. If you're building anything that needs to be fast, private, or deployed at scale, it's worth taking a look at what's now available. MiniMax-M2 is 100% open-source, free for developers right now. I have shared the link to their GitHub repo in the next tweet. You will also find the code for the playground and evaluations I've done.

Akshay 🚀

50,323 Aufrufe • vor 8 Monaten

quant at aqr sat next to me at a conference and said something that rewired how i think about markets "we don't analyze 500 stocks. we run 5 factors that explain all 500 at once" i asked him to draw it out take your entire stock universe - 500 names, 3 years of daily returns - and run PCA the 500 collapse into 5 factor 1: broad market momentum - explains 58% of all price movement by itself factor 2: sector rotation - where institutional money is actually flowing factor 3: rate sensitivity - duration risk bleeding across the equity curve factor 4: momentum tilt - which names are accelerating vs mean-reverting factor 5: credit spread stress - signal hiding inside equity vol if you know these 5 numbers, you know what 500 stocks are about to do retail watches 500 charts and sees noise. quant desks watch 5 eigenvalues and see structure math is called Principal Component Analysis, in every intro stats course on earth data to run it is free. python implementation is ~120 lines renaissance, two sigma, aqr - running this since before most retail traders could name a stock reason nobody told you isn't that it's proprietary it's that telling you costs them their edge Bookmark this market was never 500 moving parts it was always 5 - hiding inside 500
0:49

quant at aqr sat next to me at a conference and said something that rewired how i think about markets "we don't analyze 500 stocks. we run 5 factors that explain all 500 at once" i asked him to draw it out take your entire stock universe - 500 names, 3 years of daily returns - and run PCA the 500 collapse into 5 factor 1: broad market momentum - explains 58% of all price movement by itself factor 2: sector rotation - where institutional money is actually flowing factor 3: rate sensitivity - duration risk bleeding across the equity curve factor 4: momentum tilt - which names are accelerating vs mean-reverting factor 5: credit spread stress - signal hiding inside equity vol if you know these 5 numbers, you know what 500 stocks are about to do retail watches 500 charts and sees noise. quant desks watch 5 eigenvalues and see structure math is called Principal Component Analysis, in every intro stats course on earth data to run it is free. python implementation is ~120 lines renaissance, two sigma, aqr - running this since before most retail traders could name a stock reason nobody told you isn't that it's proprietary it's that telling you costs them their edge Bookmark this market was never 500 moving parts it was always 5 - hiding inside 500

Livsun

22,073 Aufrufe • vor 1 Monat