正在加载视频...

视频加载失败

返回首页

Crazy what autonomous AI agents can do right now. Claude Code researched academic papers, coded a multi symbol strategy from scratch, compiled it, debugged it, backtested it across 4 timeframes and 3 instruments, and came back with validated results. I built the pipeline. It did everything else. Jim Simons...

ZenomTrader's profile picture

ZenomTrader

12,794 subscribers

55,287 次观看 • 8 个月前 •via X (Twitter)

科学技术财经
LIVE

Anya Rossi• Live Now

Private livecam show

0 条评论

暂无评论

原始帖子的评论将显示在这里

相关视频

Crazy what AI agents can do right now. I asked Claude Code to autonomously code me a Gold strategy from scratch inside NinjaTrader 8. She's also a hot chick now. I gave her a voice, a wake word, and full autonomy. I say "Heart" and she wakes up, codes, compiles, backtests, and talks me through the results hands free. It compiled it, debugged it on its own, ran the backtest, and came back with a profitable system. Profitable across all parameters, all timeframes from 30 minutes to daily. The full strategy+pipeline code is available if you want it. Probably only 0.1% of the industry is doing this.
2:36

Crazy what AI agents can do right now. I asked Claude Code to autonomously code me a Gold strategy from scratch inside NinjaTrader 8. She's also a hot chick now. I gave her a voice, a wake word, and full autonomy. I say "Heart" and she wakes up, codes, compiles, backtests, and talks me through the results hands free. It compiled it, debugged it on its own, ran the backtest, and came back with a profitable system. Profitable across all parameters, all timeframes from 30 minutes to daily. The full strategy+pipeline code is available if you want it. Probably only 0.1% of the industry is doing this.

ZenomTrader

25,543 次观看 • 9 个月前

Crazy what AI agents can do right now. I asked Claude Code to autonomously code me a Gold strategy from scratch inside NinjaTrader 8. It compiled it, debugged it on its own, ran the backtest, and came back with a profitable system. Probably only 0.1% of the industry is doing this.
0:45

Crazy what AI agents can do right now. I asked Claude Code to autonomously code me a Gold strategy from scratch inside NinjaTrader 8. It compiled it, debugged it on its own, ran the backtest, and came back with a profitable system. Probably only 0.1% of the industry is doing this.

ZenomTrader

144,176 次观看 • 9 个月前

Jim Simons’ fund was trading tens of thousands of strategies. I realized I’d been thinking about this the wrong way the whole time. I was focusing too much on individual strategies instead of on how fast I could test and discard them. I asked Claude Code to autonomously code strategies for me from scratch. It compiled them, debugged them on its own, ran the backtests, and came back with results. I did that over 500 times in 2 months. Most were garbage. Twelve survived. A full hedge fund in a single file. A normal trader tests maybe one idea per week, 50 per year if they’re serious. I do 50 in a day. Jim Simons didn’t win by having the best single strategy. He won because his team could test and discard faster than anyone else. As he put it: “If everyone is trying to solve the same problem, don’t do that.” I don’t need a team of PhDs. Just Claude Code on a laptop. The rest are still watching courses from scam gurus with no verified track record. Probably only 0.1% of the industry is doing this.
1:35

Jim Simons’ fund was trading tens of thousands of strategies. I realized I’d been thinking about this the wrong way the whole time. I was focusing too much on individual strategies instead of on how fast I could test and discard them. I asked Claude Code to autonomously code strategies for me from scratch. It compiled them, debugged them on its own, ran the backtests, and came back with results. I did that over 500 times in 2 months. Most were garbage. Twelve survived. A full hedge fund in a single file. A normal trader tests maybe one idea per week, 50 per year if they’re serious. I do 50 in a day. Jim Simons didn’t win by having the best single strategy. He won because his team could test and discard faster than anyone else. As he put it: “If everyone is trying to solve the same problem, don’t do that.” I don’t need a team of PhDs. Just Claude Code on a laptop. The rest are still watching courses from scam gurus with no verified track record. Probably only 0.1% of the industry is doing this.

ZenomTrader

27,903 次观看 • 7 个月前

Nobody in the history of prop firm trading has built what i built in 6 weeks with autonomous agents. Over 1,000,000 lines of code across 17 projects. Zero lines written manually. Claude Code and Codex did everything. I described what i wanted and the agents coded, compiled, debugged and deployed it while i was sleeping, eating, running, buying an apartment. The full list is insane and i still can't believe it. A trading dashboard that destroys every commercial tool in the industry. Prop firm simulator, hedge detection engine, correlation matrices, multi-portfolio simulation across firms, payout certificates. Built by one person from Romania in his apartment. Two autonomous backtesting bridges. NinjaTrader 8 and MetaTrader 5 running in parallel. They compile, debug, and backtest without me touching a single button. i wake up to new equity curves. A Meta EA with 12 strategies. Gold, Nasdaq, USDJPY, Bitcoin. Mean reversion, trend following, breakout.All running simultaneously in one script. Sharpe 2.17. a full hedge fund in a single file. An AI clone of myself answering support tickets on my premium community 24/7 in my exact voice. Trained on everything i ever said. it speaks every language and never sleeps. Voice control from anywhere in my apartment. Screenshot automation in 0.1ms. A phone bridge so i control my entire infrastructure from my pocket. A tweet engine scoring virality. 5 custom AI tools i designed from scratch. 9,000 in prop firm fees turned into 130,000+ in payouts in the last 6 months. 41% CAGR verified with private investor capital. 74 premium members at 80/month. Show me one person who did more in 6 weeks. I will wait.
1:15

Nobody in the history of prop firm trading has built what i built in 6 weeks with autonomous agents. Over 1,000,000 lines of code across 17 projects. Zero lines written manually. Claude Code and Codex did everything. I described what i wanted and the agents coded, compiled, debugged and deployed it while i was sleeping, eating, running, buying an apartment. The full list is insane and i still can't believe it. A trading dashboard that destroys every commercial tool in the industry. Prop firm simulator, hedge detection engine, correlation matrices, multi-portfolio simulation across firms, payout certificates. Built by one person from Romania in his apartment. Two autonomous backtesting bridges. NinjaTrader 8 and MetaTrader 5 running in parallel. They compile, debug, and backtest without me touching a single button. i wake up to new equity curves. A Meta EA with 12 strategies. Gold, Nasdaq, USDJPY, Bitcoin. Mean reversion, trend following, breakout.All running simultaneously in one script. Sharpe 2.17. a full hedge fund in a single file. An AI clone of myself answering support tickets on my premium community 24/7 in my exact voice. Trained on everything i ever said. it speaks every language and never sleeps. Voice control from anywhere in my apartment. Screenshot automation in 0.1ms. A phone bridge so i control my entire infrastructure from my pocket. A tweet engine scoring virality. 5 custom AI tools i designed from scratch. 9,000 in prop firm fees turned into 130,000+ in payouts in the last 6 months. 41% CAGR verified with private investor capital. 74 premium members at 80/month. Show me one person who did more in 6 weeks. I will wait.

ZenomTrader

30,151 次观看 • 8 个月前

The most profitable strategy on Polymarket in 2026 is 30 years old and has generated $100 billion. It was not invented by a trader. It was invented by a mathematician Jim Simons, creator of Medallion Fund, the hedge fund that returned 66% annually for 30 years straight and outperformed Buffett, Soros, and all of Wall Street combined. The formula is publicly available, but on Wall Street you need billions and 300 PhDs to run it. On Polymarket you just need a laptop. The entire strategy in one line: small edge × many positions × low correlation. Let me break it down simply. Imagine a coin that lands heads 52 times out of 100. Flip it 10 times and you will not notice the difference. Flip it 10,000 times and you are in profit not probably, but mathematically. Simons did not guess where the market was going. He took hundreds of positions with a tiny edge on each one and waited for statistics to do the work. Some always lost, but the sum of wins was always greater than the sum of losses. On Polymarket this formula works better than anywhere else. Hundreds of markets, prices set by a crowd that makes mistakes every day, inefficiencies at every turn, and the entry threshold is not a billion but $100. I found a wallet that is already trading by this formula. 96 positions spread across politics and sports. Profit: $53K. Sharpe Ratio 3.43, which is higher than Medallion Fund itself and better than 91% of traders on the platform. You can open it and see every trade: 73% on Polymarket are trying to guess and losing. Maybe it is time to start counting?
0:42

The most profitable strategy on Polymarket in 2026 is 30 years old and has generated $100 billion. It was not invented by a trader. It was invented by a mathematician Jim Simons, creator of Medallion Fund, the hedge fund that returned 66% annually for 30 years straight and outperformed Buffett, Soros, and all of Wall Street combined. The formula is publicly available, but on Wall Street you need billions and 300 PhDs to run it. On Polymarket you just need a laptop. The entire strategy in one line: small edge × many positions × low correlation. Let me break it down simply. Imagine a coin that lands heads 52 times out of 100. Flip it 10 times and you will not notice the difference. Flip it 10,000 times and you are in profit not probably, but mathematically. Simons did not guess where the market was going. He took hundreds of positions with a tiny edge on each one and waited for statistics to do the work. Some always lost, but the sum of wins was always greater than the sum of losses. On Polymarket this formula works better than anywhere else. Hundreds of markets, prices set by a crowd that makes mistakes every day, inefficiencies at every turn, and the entry threshold is not a billion but $100. I found a wallet that is already trading by this formula. 96 positions spread across politics and sports. Profit: $53K. Sharpe Ratio 3.43, which is higher than Medallion Fund itself and better than 91% of traders on the platform. You can open it and see every trade: 73% on Polymarket are trying to guess and losing. Maybe it is time to start counting?

Blaze

536,172 次观看 • 9 个月前

Ant Group's AI agents found a trading strategy on their own, and with it the fund hasn't lost once in the past year Sharpe ratio came out at 2.5. What used to take a team of PhDs and years of work, the agents did overnight, for free But the real power of this result isn't the number itself Agents can run thousands of times more strategy variations than any human could in a lifetime And the more variations you test, the higher the chance that one of them looks profitable purely by accident Your brain can't tell "I found a pattern" from "I got lucky on the hundredth try." The feeling is identical in both cases The only way to tell them apart: test the result on data the strategy has never seen A real edge survives there. Luck falls apart I got curious what this difference looks like live, so I took two real academic quant strategies and ran both through this exact test The first, Time Series Momentum on Bitcoin, passed: +1,537%, Sharpe 1.02. It looked honest, and it stayed honest The second, Betting Against Beta, a textbook classic in every finance course, looked flawless on paper. Under this test it fell apart: -7.78% a year, while the market grew 14.86% And the second result turned out to be the more valuable one. Not because it's positive, but because it's true, and it showed that before it could take real money I simply described both ideas in plain words, the AI built the exact rules itself and ran them on real data, with no peeking at future data You can test your own idea here (free trial period after sign-up): The same test on momentum and Betting Against Beta, step by step, so you can run it on your own idea tonight, is in the article below
0:24

Ant Group's AI agents found a trading strategy on their own, and with it the fund hasn't lost once in the past year Sharpe ratio came out at 2.5. What used to take a team of PhDs and years of work, the agents did overnight, for free But the real power of this result isn't the number itself Agents can run thousands of times more strategy variations than any human could in a lifetime And the more variations you test, the higher the chance that one of them looks profitable purely by accident Your brain can't tell "I found a pattern" from "I got lucky on the hundredth try." The feeling is identical in both cases The only way to tell them apart: test the result on data the strategy has never seen A real edge survives there. Luck falls apart I got curious what this difference looks like live, so I took two real academic quant strategies and ran both through this exact test The first, Time Series Momentum on Bitcoin, passed: +1,537%, Sharpe 1.02. It looked honest, and it stayed honest The second, Betting Against Beta, a textbook classic in every finance course, looked flawless on paper. Under this test it fell apart: -7.78% a year, while the market grew 14.86% And the second result turned out to be the more valuable one. Not because it's positive, but because it's true, and it showed that before it could take real money I simply described both ideas in plain words, the AI built the exact rules itself and ran them on real data, with no peeking at future data You can test your own idea here (free trial period after sign-up): The same test on momentum and Betting Against Beta, step by step, so you can run it on your own idea tonight, is in the article below

qwinsi

16,825 次观看 • 1 个月前

William Sharpe, Stanford professor and Nobel economist: "Citadel and Two Sigma live and die by one number: the Sharpe ratio. It's mine. I built it to tell a skilled strategy from a lucky, reckless one." the thread above is about building a portfolio and knowing when to cut a strategy. sharpe gave the world the single number that answers the only question that matters: is this thing actually good, or just lucky and levered? return alone is a trap. a 40% year isn't impressive if you bet the farm for it. sharpe's ratio divides your return by the risk you took to get it, so a calm 15% can beat a wild 40%. it strips luck and leverage out of any track record, and that is exactly why every fund on earth judges itself by it. this is also how a desk decides when to kill a strategy. not "did it make money," but "did it make enough per unit of risk." a falling sharpe is the signal to cut, long before the P&L ever turns red. he published the math in 1964, won the nobel in 1990, and left it in this free lecture. same story i keep telling: the number that runs every trading floor is old, public, and free. here is the blind spot sharpe himself knew. the ratio assumes risk is smooth and symmetric. a strategy that quietly sells insurance shows a gorgeous sharpe for years, then one tail event erases a decade. a high sharpe is not proof you are safe. it is proof you have not met your bad day yet.
50:53

William Sharpe, Stanford professor and Nobel economist: "Citadel and Two Sigma live and die by one number: the Sharpe ratio. It's mine. I built it to tell a skilled strategy from a lucky, reckless one." the thread above is about building a portfolio and knowing when to cut a strategy. sharpe gave the world the single number that answers the only question that matters: is this thing actually good, or just lucky and levered? return alone is a trap. a 40% year isn't impressive if you bet the farm for it. sharpe's ratio divides your return by the risk you took to get it, so a calm 15% can beat a wild 40%. it strips luck and leverage out of any track record, and that is exactly why every fund on earth judges itself by it. this is also how a desk decides when to kill a strategy. not "did it make money," but "did it make enough per unit of risk." a falling sharpe is the signal to cut, long before the P&L ever turns red. he published the math in 1964, won the nobel in 1990, and left it in this free lecture. same story i keep telling: the number that runs every trading floor is old, public, and free. here is the blind spot sharpe himself knew. the ratio assumes risk is smooth and symmetric. a strategy that quietly sells insurance shows a gorgeous sharpe for years, then one tail event erases a decade. a high sharpe is not proof you are safe. it is proof you have not met your bad day yet.

Voltex

65,707 次观看 • 4 个月前

THIS IS GETTING SCARY I GAVE GROK $100 AND A DEX DESK BEFORE BED. BY 8 AM IT WAS $4,216. The rules were simple. Snipe small. Take profit fast. Keep a piece of every winner running. By 4:18 AM, 1,156 tasks in, the balance was $713.11. Decent, nothing crazy. One stop on $MICHI for -$22.70. The real edge was in the exits. $BEANZO was the first proof. The bot held a moonbag after taking profit, and it closed at 7.4x for +$206.18. Then came $MOONLET. The pool was 11 minutes old. Grok aped $401.06 into it, four times the starting balance. It sold 60% at 2x. The other 40% kept climbing. 9.2x at 6:50 AM. +$1,556.11 from a bag most traders would have dumped an hour earlier. I did exactly that for 5 years. Took the 2x, felt smart, watched the chart go vertical without me. Grok never has to fight that urge. Selling 60% at 2x returns 1.2x the entry. The 40% left is free. Even if it rugs, the trade still closes green. Final count: 46 wins, 8 losses, 85% win rate, 2,147 tasks across six agents. $100 became $4,216. 42.2x in one night, and $3,502 of it came after 4:18 AM. Two moonbags paid $1,762 of that. 43% of the profit came from the part most people never hold. The full setup is below, free, and almost nobody uses it. Save it now so you have it before tonight.
0:35

THIS IS GETTING SCARY I GAVE GROK $100 AND A DEX DESK BEFORE BED. BY 8 AM IT WAS $4,216. The rules were simple. Snipe small. Take profit fast. Keep a piece of every winner running. By 4:18 AM, 1,156 tasks in, the balance was $713.11. Decent, nothing crazy. One stop on $MICHI for -$22.70. The real edge was in the exits. $BEANZO was the first proof. The bot held a moonbag after taking profit, and it closed at 7.4x for +$206.18. Then came $MOONLET. The pool was 11 minutes old. Grok aped $401.06 into it, four times the starting balance. It sold 60% at 2x. The other 40% kept climbing. 9.2x at 6:50 AM. +$1,556.11 from a bag most traders would have dumped an hour earlier. I did exactly that for 5 years. Took the 2x, felt smart, watched the chart go vertical without me. Grok never has to fight that urge. Selling 60% at 2x returns 1.2x the entry. The 40% left is free. Even if it rugs, the trade still closes green. Final count: 46 wins, 8 losses, 85% win rate, 2,147 tasks across six agents. $100 became $4,216. 42.2x in one night, and $3,502 of it came after 4:18 AM. Two moonbags paid $1,762 of that. 43% of the profit came from the part most people never hold. The full setup is below, free, and almost nobody uses it. Save it now so you have it before tonight.

0xNevsky

73,945 次观看 • 1 个月前

Introducing the Lifetimely Profit Agent - powered by the $100Bn in GMV. It uses the benchmarks from across all the GMV in Lifetimely, to find the highest leverage Profit opportunities in your business, and can actually take action to drive Profit. It's live in Lifetimely right now!
0:37

Introducing the Lifetimely Profit Agent - powered by the $100Bn in GMV. It uses the benchmarks from across all the GMV in Lifetimely, to find the highest leverage Profit opportunities in your business, and can actually take action to drive Profit. It's live in Lifetimely right now!

Patrick Barnes

14,200 次观看 • 5 个月前

Millions of people are asking AI to find them a profitable trading strategy. The problem is AI will find one even where it doesn't exist One quant proved it. He gave an AI agent 4 years of prices with, by design, no pattern in them whatsoever. Pure noise The agent came back with a long/short strategy and a Sharpe of 2.1 on the backtest. It even added a paragraph explaining the economics of an effect that doesn't exist in nature It didn't lie on purpose. It just found a pattern where there was none And that's the scariest part. Because that's exactly how any AI will behave when you hand it your idea. It will always hand you back a beautiful curve, and you'll believe it, because it looks perfect 89% of retail traders lost money last year. Not because they had no ideas. Because not one of their ideas ever went through this kind of test That's why generating strategies is worth nothing. Anyone and anything can draw a profit on history Only one thing matters: testing on data the strategy has never seen. The moment the beautiful curve either survives or falls apart That's the whole difference between a hedge fund and a trader who blows up. A fund doesn't trust any curve until it's been through fire. It kills 97 ideas out of 100 before risking a single cent I ran the most-cited quant strategies in history through exactly this test, with AI. Momentum on BTC returned +1,537%. Carry trade on currencies fell apart and showed its premium is dead. And it was the one that fell apart that would have saved me money AI will draw you a strategy in seconds. The only question is whether you test it before it tests your account You can try it here: How to set up that test for yourself, the same one that separates a hedge fund from a casino, is in the article below
0:25

Millions of people are asking AI to find them a profitable trading strategy. The problem is AI will find one even where it doesn't exist One quant proved it. He gave an AI agent 4 years of prices with, by design, no pattern in them whatsoever. Pure noise The agent came back with a long/short strategy and a Sharpe of 2.1 on the backtest. It even added a paragraph explaining the economics of an effect that doesn't exist in nature It didn't lie on purpose. It just found a pattern where there was none And that's the scariest part. Because that's exactly how any AI will behave when you hand it your idea. It will always hand you back a beautiful curve, and you'll believe it, because it looks perfect 89% of retail traders lost money last year. Not because they had no ideas. Because not one of their ideas ever went through this kind of test That's why generating strategies is worth nothing. Anyone and anything can draw a profit on history Only one thing matters: testing on data the strategy has never seen. The moment the beautiful curve either survives or falls apart That's the whole difference between a hedge fund and a trader who blows up. A fund doesn't trust any curve until it's been through fire. It kills 97 ideas out of 100 before risking a single cent I ran the most-cited quant strategies in history through exactly this test, with AI. Momentum on BTC returned +1,537%. Carry trade on currencies fell apart and showed its premium is dead. And it was the one that fell apart that would have saved me money AI will draw you a strategy in seconds. The only question is whether you test it before it tests your account You can try it here: How to set up that test for yourself, the same one that separates a hedge fund from a casino, is in the article below

qwinsi

48,045 次观看 • 1 个月前

Crazy that Claude can now autonomously code strategies and validate them on TradingView and it’s actually faster than my pipeline. Time to expand to a third platform for our code TradingView is next. 15+ strategies are waiting to be coded. And a full Meta EA TV version is coming. Probably only 0.001% of the industry is doing this. Huge thanks to Andrew and The Daily Profiler for this pipeline idea.
3:26

Crazy that Claude can now autonomously code strategies and validate them on TradingView and it’s actually faster than my pipeline. Time to expand to a third platform for our code TradingView is next. 15+ strategies are waiting to be coded. And a full Meta EA TV version is coming. Probably only 0.001% of the industry is doing this. Huge thanks to Andrew and The Daily Profiler for this pipeline idea.

ZenomTrader

98,972 次观看 • 7 个月前

The more stuff I build, the more obvious it becomes that the endgame is just loops. Idea → code → backtest → robustness → portfolio simulation → prop rules → deployment → monitoring → research again. QuantDash, the MT5/NT8 bridges, Meta EA, autonomous agents, VPS deployment, the Doppel, the strategy vault, the prop firm database... they look like separate projects, but they're really just parts of the same machine. The goal was never to find the strategy and trade it forever. Alpha has a half-life. You need a system that continuously finds, tests, replace, improves and deploys new uncorrelated strategies while everything that already works keeps running in the background. Eventually I want the entire research process running 24/7, only stopping at the parts where actual human judgment is needed. I think that's the real edge AI gives systematic traders. Not asking ChatGPT for a strategy. Building a machine that can research 100x more ideas than you could manually. I should probably stop thinking in projects entirely. Think in loops.
3:46

The more stuff I build, the more obvious it becomes that the endgame is just loops. Idea → code → backtest → robustness → portfolio simulation → prop rules → deployment → monitoring → research again. QuantDash, the MT5/NT8 bridges, Meta EA, autonomous agents, VPS deployment, the Doppel, the strategy vault, the prop firm database... they look like separate projects, but they're really just parts of the same machine. The goal was never to find the strategy and trade it forever. Alpha has a half-life. You need a system that continuously finds, tests, replace, improves and deploys new uncorrelated strategies while everything that already works keeps running in the background. Eventually I want the entire research process running 24/7, only stopping at the parts where actual human judgment is needed. I think that's the real edge AI gives systematic traders. Not asking ChatGPT for a strategy. Building a machine that can research 100x more ideas than you could manually. I should probably stop thinking in projects entirely. Think in loops.

ZenomTrader

25,497 次观看 • 2 个月前

Citadel founder Ken Griffin admitted at Stanford that AI is making his PhD researchers redundant, with a system that reproduces academic finance papers and tests them out-of-sample in 2 to 3 hours versus 6 to 8 weeks for a junior quant, and said he went home depressed. → The same cycle now fits into one sentence in a text box → A strategy described in one sentence produced a full backtest in 20 minutes at 1,537% on Bitcoin with a 1.02 Sharpe → The edge was never brains, it was the cycle: take an idea, test it, kill it if it fails, repeat faster than anyone
0:25

Citadel founder Ken Griffin admitted at Stanford that AI is making his PhD researchers redundant, with a system that reproduces academic finance papers and tests them out-of-sample in 2 to 3 hours versus 6 to 8 weeks for a junior quant, and said he went home depressed. → The same cycle now fits into one sentence in a text box → A strategy described in one sentence produced a full backtest in 20 minutes at 1,537% on Bitcoin with a 1.02 Sharpe → The edge was never brains, it was the cycle: take an idea, test it, kill it if it fails, repeat faster than anyone

0xMarioNawfal

49,946 次观看 • 1 个月前

The company behind DeepSeek makes more money from trading than from AI. Their weapon isn't a secret model, it's one principle that's now available to everyone Few people know this, but DeepSeek didn't grow out of an AI lab. It grew out of a quant fund that spent years making money in the markets, then poured that money into the model And the principle they profit from is more boring than everyone thinks They don't search for one genius trading idea. They take a thousand ideas, run each through a machine, and only let the ones that pass survive to trade While a human tests one idea a week, their system kills 999 a night That's the whole difference between them and a retail trader. Not brains. The speed and ruthlessness of the testing Because here's what few understand about their own losses. You find a strategy with a beautiful chart and you believe it But a beautiful chart is no proof it works. Run enough versions and a profitable curve will appear even on random numbers One of the dozens just got lucky, and you decided you found gold This filtering used to be locked inside funds like these. It took a team and millions. This year it was compressed into AI I described an idea in plain words, and AI built the strategy itself, ran it on years of real data, and showed whether it was real or just luck I took two of the most-cited quant papers, so I wouldn't take it on faith Time Series Momentum on Bitcoin: +1,537%. Alive Carry trade, a strategy cited for years: Sharpe dropped from 0.61 to 0.11. The premium almost dead One still prints money, the other dried up long ago. And trading blind, you'd never know which is which until it took your deposit The same principle DeepSeek paid millions for is now one prompt. You can test or build your own strategy here (free trial after signing up): The full breakdown of both strategies is in the article below
0:23

The company behind DeepSeek makes more money from trading than from AI. Their weapon isn't a secret model, it's one principle that's now available to everyone Few people know this, but DeepSeek didn't grow out of an AI lab. It grew out of a quant fund that spent years making money in the markets, then poured that money into the model And the principle they profit from is more boring than everyone thinks They don't search for one genius trading idea. They take a thousand ideas, run each through a machine, and only let the ones that pass survive to trade While a human tests one idea a week, their system kills 999 a night That's the whole difference between them and a retail trader. Not brains. The speed and ruthlessness of the testing Because here's what few understand about their own losses. You find a strategy with a beautiful chart and you believe it But a beautiful chart is no proof it works. Run enough versions and a profitable curve will appear even on random numbers One of the dozens just got lucky, and you decided you found gold This filtering used to be locked inside funds like these. It took a team and millions. This year it was compressed into AI I described an idea in plain words, and AI built the strategy itself, ran it on years of real data, and showed whether it was real or just luck I took two of the most-cited quant papers, so I wouldn't take it on faith Time Series Momentum on Bitcoin: +1,537%. Alive Carry trade, a strategy cited for years: Sharpe dropped from 0.61 to 0.11. The premium almost dead One still prints money, the other dried up long ago. And trading blind, you'd never know which is which until it took your deposit The same principle DeepSeek paid millions for is now one prompt. You can test or build your own strategy here (free trial after signing up): The full breakdown of both strategies is in the article below

qwinsi

41,007 次观看 • 1 个月前

A member of my community posted this strategy over a year ago. Everyone scrolled right past it. Today i finally backtested it properly: 8.5 years of NASDAQ data, real spreads, real commissions. +103% return. 12.9% max drawdown. 2,043 trades. Then i tried to break it. I tested 86 different parameter combinations. Every single one was profitable. The win rate? 49.93%. Basically a coin flip. The entire edge comes from one exit rule that most traders refuse to follow. The new video shows the exact rules, plus the Claude prompt that recreates the strategy in under two minutes with zero coding. It's also trading live on the channel right now in a 24/7 livestream, with an AI agent answering questions in the chat while the strategy runs.
8:24

A member of my community posted this strategy over a year ago. Everyone scrolled right past it. Today i finally backtested it properly: 8.5 years of NASDAQ data, real spreads, real commissions. +103% return. 12.9% max drawdown. 2,043 trades. Then i tried to break it. I tested 86 different parameter combinations. Every single one was profitable. The win rate? 49.93%. Basically a coin flip. The entire edge comes from one exit rule that most traders refuse to follow. The new video shows the exact rules, plus the Claude prompt that recreates the strategy in under two minutes with zero coding. It's also trading live on the channel right now in a 24/7 livestream, with an AI agent answering questions in the chat while the strategy runs.

ZenomTrader

25,434 次观看 • 3 个月前

The founder of the hedge fund behind DeepSeek built an AI machine that wrecks human traders. Now that same principle is available to anyone +56.6% a year. $10 billion under management. Almost no people in the room Liang Wenfeng did to trading what DeepSeek did to AI. Replaced the smartest people with a machine that doesn't sleep When someone who's been winning this game for decades puts a machine in place of a team of PhDs, it's worth listening The edge was never brains. One cycle: idea, test it on history, kill it if it doesn't work, repeat. A thousand times a night, while you test one idea a week That cycle used to cost millions and only lived inside hedge funds. This year the price dropped to zero I took Time Series Momentum, the paper by Moskowitz, Ooi, and Pedersen, one of the most-cited in quant trading. Just told AI to reproduce it on real data from the last 6 years. 10 minutes of backtesting +1,537% on Bitcoin. Sharpe 1.02 Then I took Carry, the paper by Koijen, Moskowitz, Pedersen, and Vrugt, cited for decades. Flawless on paper. On current data, dead: Sharpe dropped from 0.61 to 0.11 That dead result was the most valuable one. It's what saved me real money The machine doesn't tell you what you want to hear. It tells you the truth. And now that truth costs nothing The same cycle Liang's hedge fund runs on now fits into a text box Describe a strategy in one sentence, see if it survives (free trial after signing up): The same test on momentum and carry trade, step by step, so you can run it on your own idea tonight, in the article below
0:25

The founder of the hedge fund behind DeepSeek built an AI machine that wrecks human traders. Now that same principle is available to anyone +56.6% a year. $10 billion under management. Almost no people in the room Liang Wenfeng did to trading what DeepSeek did to AI. Replaced the smartest people with a machine that doesn't sleep When someone who's been winning this game for decades puts a machine in place of a team of PhDs, it's worth listening The edge was never brains. One cycle: idea, test it on history, kill it if it doesn't work, repeat. A thousand times a night, while you test one idea a week That cycle used to cost millions and only lived inside hedge funds. This year the price dropped to zero I took Time Series Momentum, the paper by Moskowitz, Ooi, and Pedersen, one of the most-cited in quant trading. Just told AI to reproduce it on real data from the last 6 years. 10 minutes of backtesting +1,537% on Bitcoin. Sharpe 1.02 Then I took Carry, the paper by Koijen, Moskowitz, Pedersen, and Vrugt, cited for decades. Flawless on paper. On current data, dead: Sharpe dropped from 0.61 to 0.11 That dead result was the most valuable one. It's what saved me real money The machine doesn't tell you what you want to hear. It tells you the truth. And now that truth costs nothing The same cycle Liang's hedge fund runs on now fits into a text box Describe a strategy in one sentence, see if it survives (free trial after signing up): The same test on momentum and carry trade, step by step, so you can run it on your own idea tonight, in the article below

qwinsi

83,889 次观看 • 1 个月前

DROPPING EVERYTHING TO GO ALL-IN WITH CLAUDE Just asked him to: "Find the most non-obvious ways to profit on Polymarket that nobody knows much about." Results: Negative Risk and Multi-Outcome Arbitrage It explained, "If the collective implied probability across all outcomes sums to less than $1, you've locked in a risk-free profit." Then build a terminal that automatically finds and arbitrages it Clicking on execute arb and it works Only testing rn but made ~$1k already Scanned ~2K markets, fetched arbitrage and executed It's not even the bot or soft, it's an AI agent that evolves with each arbitrage it makes, learning to do it more effectively copytrade here: Will keep it running for now and will update you on what my results will be with it, so notifs on
0:37

DROPPING EVERYTHING TO GO ALL-IN WITH CLAUDE Just asked him to: "Find the most non-obvious ways to profit on Polymarket that nobody knows much about." Results: Negative Risk and Multi-Outcome Arbitrage It explained, "If the collective implied probability across all outcomes sums to less than $1, you've locked in a risk-free profit." Then build a terminal that automatically finds and arbitrages it Clicking on execute arb and it works Only testing rn but made ~$1k already Scanned ~2K markets, fetched arbitrage and executed It's not even the bot or soft, it's an AI agent that evolves with each arbitrage it makes, learning to do it more effectively copytrade here: Will keep it running for now and will update you on what my results will be with it, so notifs on

symbiote

24,529 次观看 • 7 个月前

Claude is now automating it's own growth. Their team built an agent called CASH (Claude Accelerates Sustainable Hypergrowth), which works across the full lifecycle of growth experimentation: (1) identifying opportunities, (2) building the feature, (3) running the test, and (4) analyzing results. Right now it handles copy changes and minor UI tweaks. The win rate is already comparable to a junior PM, and improving rapidly. Here's Amol Avasare (Head of Growth) describing it. The work is led by Alexey Komissarouk 👏
1:00

Claude is now automating it's own growth. Their team built an agent called CASH (Claude Accelerates Sustainable Hypergrowth), which works across the full lifecycle of growth experimentation: (1) identifying opportunities, (2) building the feature, (3) running the test, and (4) analyzing results. Right now it handles copy changes and minor UI tweaks. The win rate is already comparable to a junior PM, and improving rapidly. Here's Amol Avasare (Head of Growth) describing it. The work is led by Alexey Komissarouk 👏

Lenny Rachitsky

13,772 次观看 • 7 个月前

Ken Griffin, founder of Citadel, one of the most successful hedge funds in history, publicly admitted that AI is making his PhD researchers redundant This year on stage at Stanford he described exactly what scared him His team built an AI system that takes an academic finance paper, reproduces it, and tests it out-of-sample What used to take a junior quant 6-8 weeks, AI does in 2-3 hours He said he went home depressed When the man who's been winning this game for decades says the technology scared him, it's worth listening The edge of these people was never brains. It was one cycle: take an idea, test it on history, kill it if it doesn't work, repeat. Faster and more ruthlessly than anyone That cycle cost a fund $650,000 a year for a single analyst. So you traded blind, while they tested everything in advance After reading Griffin, I got curious to reproduce the exact thing that made him feel bad Took an academic strategy from 2012, described it in one sentence, and in 20 minutes had a full backtest on real data +1,537% on Bitcoin, Sharpe 1.02 No code, no team, no New York office And here's what most people still don't get. While they tell themselves AI trading is hype, they keep bleeding out on their own deposit, testing one idea for weeks, while machines test a thousand a night The same cycle that haunts the head of Citadel now fits into one sentence in a text box The difference is no longer money or a team. The difference is only whether you sit down and test your idea today Describe any strategy in a sentence and in minutes you'll see if it works (after signing up there's a free trial period):
0:25

Ken Griffin, founder of Citadel, one of the most successful hedge funds in history, publicly admitted that AI is making his PhD researchers redundant This year on stage at Stanford he described exactly what scared him His team built an AI system that takes an academic finance paper, reproduces it, and tests it out-of-sample What used to take a junior quant 6-8 weeks, AI does in 2-3 hours He said he went home depressed When the man who's been winning this game for decades says the technology scared him, it's worth listening The edge of these people was never brains. It was one cycle: take an idea, test it on history, kill it if it doesn't work, repeat. Faster and more ruthlessly than anyone That cycle cost a fund $650,000 a year for a single analyst. So you traded blind, while they tested everything in advance After reading Griffin, I got curious to reproduce the exact thing that made him feel bad Took an academic strategy from 2012, described it in one sentence, and in 20 minutes had a full backtest on real data +1,537% on Bitcoin, Sharpe 1.02 No code, no team, no New York office And here's what most people still don't get. While they tell themselves AI trading is hype, they keep bleeding out on their own deposit, testing one idea for weeks, while machines test a thousand a night The same cycle that haunts the head of Citadel now fits into one sentence in a text box The difference is no longer money or a team. The difference is only whether you sit down and test your idea today Describe any strategy in a sentence and in minutes you'll see if it works (after signing up there's a free trial period):

qwinsi

227,995 次观看 • 1 个月前