Загрузка видео...

Не удалось загрузить видео

На главную

an ex-OpenAI researcher pushed his trading system to GitHub the night he got fired someone in a crypto Discord dropped the link at 2am. "this repo appeared an hour ago. zero stars. readme says internal use only. the code looks real" i clicked. it was a complete prediction market...

hanakoxbt's profile picture

Hanako

9,715 subscribers

23,546 просмотров • 2 месяцев назад •via X (Twitter)

Новости и политикаНаука и технологииФинансы
LIVE

Anya Rossi• Live Now

Private livecam show

Комментарии: 0

Нет доступных комментариев

Здесь появятся комментарии из оригинального поста

Похожие видео

a Renaissance intern posted his model on GitHub for 6 hours before deleting it someone in a Discord dropped the link at 1am. "this looks real. star count is zero. repo is 4 hours old. readme has internal formatting" i clicked. it was a full prediction market scoring model. four filters: > disposition ratio > category dominance > capital velocity > volume exit triggers clean Python. commented like someone who writes for a team. the readme said "internal use - do not distribute" i ran git clone so fast i typo'd the first attempt. by 3am the repo was gone. mass-deleted. commits. branches. everything. but i had it. opened Claude. gave it the code and one public repo: 2,600 stars. official framework. LLM-powered trading agents. "merge this scoring model with this execution layer. build a terminal that runs all four filters on live markets" Claude had it running by sunrise. first three days i was terrified it would break. kept checking every hour. it didn't break. it just kept scoring and entering. > AMD Xilinx - entered 52c. model said 59c. closed +7c in 2h. > Artemis launch - entered 63c. model said 85c. closed +22c in 5h. > Derecho MW - entered 71c. model said 87c. closed +16c in under 2h. disposition filter kills everything below 0.70 automatically. no override. every wallet it copies is filtered to one category. crypto wallet copies crypto. weather wallet copies weather. nothing crosses. 87% win rate. 34 markets. +$12,685. 6 strategies running simultaneously: base rate. conditional prob. liquidity snipe. delta hedge. calibration. theta decay. spread arb catching gaps live: - BTC 92K vs Kalshi gap 0.3c - Fed cut vs PredictIt gap 3.7c - Derecho vs Betfair gap 11.9c copytrade setup here: cost: Claude and a VPS. cheaper than the mass-deleted repo was worth. i checked his GitHub profile last week. still active. new repos. normal stuff. he has no idea someone cloned it. and now you're reading what was in it.
0:40

a Renaissance intern posted his model on GitHub for 6 hours before deleting it someone in a Discord dropped the link at 1am. "this looks real. star count is zero. repo is 4 hours old. readme has internal formatting" i clicked. it was a full prediction market scoring model. four filters: > disposition ratio > category dominance > capital velocity > volume exit triggers clean Python. commented like someone who writes for a team. the readme said "internal use - do not distribute" i ran git clone so fast i typo'd the first attempt. by 3am the repo was gone. mass-deleted. commits. branches. everything. but i had it. opened Claude. gave it the code and one public repo: 2,600 stars. official framework. LLM-powered trading agents. "merge this scoring model with this execution layer. build a terminal that runs all four filters on live markets" Claude had it running by sunrise. first three days i was terrified it would break. kept checking every hour. it didn't break. it just kept scoring and entering. > AMD Xilinx - entered 52c. model said 59c. closed +7c in 2h. > Artemis launch - entered 63c. model said 85c. closed +22c in 5h. > Derecho MW - entered 71c. model said 87c. closed +16c in under 2h. disposition filter kills everything below 0.70 automatically. no override. every wallet it copies is filtered to one category. crypto wallet copies crypto. weather wallet copies weather. nothing crosses. 87% win rate. 34 markets. +$12,685. 6 strategies running simultaneously: base rate. conditional prob. liquidity snipe. delta hedge. calibration. theta decay. spread arb catching gaps live: - BTC 92K vs Kalshi gap 0.3c - Fed cut vs PredictIt gap 3.7c - Derecho vs Betfair gap 11.9c copytrade setup here: cost: Claude and a VPS. cheaper than the mass-deleted repo was worth. i checked his GitHub profile last week. still active. new repos. normal stuff. he has no idea someone cloned it. and now you're reading what was in it.

Hanako

117,572 просмотров • 3 месяцев назад

a Chinese PhD dropout told me Wall Street is solving the wrong equation on purpose i met him through a girl i was seeing in hong kong. he hadn't had a real job in three years. "every bank. every fund. every quant lab. they all try to predict where BTC goes next. up or down. that's the wrong question" i asked what the right question is. "who is mispricing the reaction" he opened his phone. showed me Polymarket. "288 BTC markets per day. every 5 minutes. YES or NO. the crowd bets direction. direction doesn't matter. what matters is the spread between YES and NO when the crowd panics" he drew on a napkin. "YES at 79c. NO at 22c. combined $1.01. fair. then someone dumps $3,000 on YES. it spikes to 96c. NO crashes to 5c. that's 17 cents of free alpha sitting there for 4 minutes" "Wall Street builds billion dollar models to predict direction. i buy the panic. different equation. different result" i flew home. opened a Chinese AI. typed what he told me. "build a spread capture engine for Polymarket 5-minute BTC markets. buy the underpriced side when crowd panic creates a gap. never predict direction" 30,800 stars. multi-agent LLM trading framework. the AI adapted it for prediction markets overnight. NERVE GRID. 6 pairs. BTC. ETH. SOL. AVAX. 400+ windows per day. edge table live: BTC+100K 5m. 79c vs 22c. +27c edge. BTC<99K. 35c vs 27c. +8c edge. ETH+3.8K. 46c vs 30c. +12c edge. SOL+200 5m. 55c vs 22c. +22c edge. the system kills 2,986 setups per day. enters 563. avg hold 1:02. wins are 4.51x the losses. fill rate 96.4%. avg latency 15ms. best fill 9ms. +$4,847 from $600 seed. infrastructure: $20 AI. $5 VPS. everything else open source. $25/month. i copied the setup here: texted him two weeks later with a screenshot. "you're running the panic equation?" the AI built it from one conversation. "good. now try ETH and SOL. bigger spreads. less competition" i already was. he could see it in the screenshot. long pause. "you learn fast for someone who asks the right questions" i didn't ask any questions. i just listened.
0:50

a Chinese PhD dropout told me Wall Street is solving the wrong equation on purpose i met him through a girl i was seeing in hong kong. he hadn't had a real job in three years. "every bank. every fund. every quant lab. they all try to predict where BTC goes next. up or down. that's the wrong question" i asked what the right question is. "who is mispricing the reaction" he opened his phone. showed me Polymarket. "288 BTC markets per day. every 5 minutes. YES or NO. the crowd bets direction. direction doesn't matter. what matters is the spread between YES and NO when the crowd panics" he drew on a napkin. "YES at 79c. NO at 22c. combined $1.01. fair. then someone dumps $3,000 on YES. it spikes to 96c. NO crashes to 5c. that's 17 cents of free alpha sitting there for 4 minutes" "Wall Street builds billion dollar models to predict direction. i buy the panic. different equation. different result" i flew home. opened a Chinese AI. typed what he told me. "build a spread capture engine for Polymarket 5-minute BTC markets. buy the underpriced side when crowd panic creates a gap. never predict direction" 30,800 stars. multi-agent LLM trading framework. the AI adapted it for prediction markets overnight. NERVE GRID. 6 pairs. BTC. ETH. SOL. AVAX. 400+ windows per day. edge table live: BTC+100K 5m. 79c vs 22c. +27c edge. BTC<99K. 35c vs 27c. +8c edge. ETH+3.8K. 46c vs 30c. +12c edge. SOL+200 5m. 55c vs 22c. +22c edge. the system kills 2,986 setups per day. enters 563. avg hold 1:02. wins are 4.51x the losses. fill rate 96.4%. avg latency 15ms. best fill 9ms. +$4,847 from $600 seed. infrastructure: $20 AI. $5 VPS. everything else open source. $25/month. i copied the setup here: texted him two weeks later with a screenshot. "you're running the panic equation?" the AI built it from one conversation. "good. now try ETH and SOL. bigger spreads. less competition" i already was. he could see it in the screenshot. long pause. "you learn fast for someone who asks the right questions" i didn't ask any questions. i just listened.

Hanako

37,205 просмотров • 2 месяцев назад

An OpenAI researcher sat down next to me at a coffee shop in Mission District I had my terminal open. Three panels. Live trades scrolling. He was reading something on his laptop. Glanced over. Stopped reading. "That's not a dashboard. That's a live scoring engine. What model is running that" I told him. Claude Code. Four repos. $25 a month. He closed his laptop. "I work at OpenAI. We benchmarked Claude internally last month. You're using it to trade prediction markets?" I opened one link. 86 million trades. Every wallet. Every entry. Every exit. The entire Polymarket history since day one. "This is public? We quoted a seven-figure budget to reconstruct this kind of dataset from on-chain data. The project is still in review" I told him Claude Code connects directly. It reads the whole dataset. Finds the wallets that win. Then finds WHY they win. Then copies the pattern. He pulled his chair closer. "Walk me through the exit logic" Top wallets exit before resolution 91% of the time. They capture 86% of the move and cut losers at 12%. Everyone else holds to 58%. Same entries. Completely different exits. My bot cuts at 85% of expected move. Or on a 3x volume spike. Whichever hits first. "Who gave you that threshold" Claude Code found it in poly_data. In about 20 minutes. "We had a team of nine working on this exact problem for six months. They never shipped it. You did it in a weekend with a competitor's model" I opened another link. Three commands. 500+ markets. No API key. Claude scores them in 20 minutes. "That's our internal eval pipeline. Except it took us six months and you built it on a Saturday" My setup: Claude API - $20/mo VPS - $5/mo poly_data - free polymarket-cli - free 19 days. 4 agents. 74% win rate. +$9,400. Copytrade here: I showed him the article where I broke down every repo, every command, every dollar. He read it for five minutes. Then looked up. "You just published what we presented to Sam last quarter. Using the other team's model" He texted me the next morning. "My director found your thread. Take it down" Too late.
0:20

An OpenAI researcher sat down next to me at a coffee shop in Mission District I had my terminal open. Three panels. Live trades scrolling. He was reading something on his laptop. Glanced over. Stopped reading. "That's not a dashboard. That's a live scoring engine. What model is running that" I told him. Claude Code. Four repos. $25 a month. He closed his laptop. "I work at OpenAI. We benchmarked Claude internally last month. You're using it to trade prediction markets?" I opened one link. 86 million trades. Every wallet. Every entry. Every exit. The entire Polymarket history since day one. "This is public? We quoted a seven-figure budget to reconstruct this kind of dataset from on-chain data. The project is still in review" I told him Claude Code connects directly. It reads the whole dataset. Finds the wallets that win. Then finds WHY they win. Then copies the pattern. He pulled his chair closer. "Walk me through the exit logic" Top wallets exit before resolution 91% of the time. They capture 86% of the move and cut losers at 12%. Everyone else holds to 58%. Same entries. Completely different exits. My bot cuts at 85% of expected move. Or on a 3x volume spike. Whichever hits first. "Who gave you that threshold" Claude Code found it in poly_data. In about 20 minutes. "We had a team of nine working on this exact problem for six months. They never shipped it. You did it in a weekend with a competitor's model" I opened another link. Three commands. 500+ markets. No API key. Claude scores them in 20 minutes. "That's our internal eval pipeline. Except it took us six months and you built it on a Saturday" My setup: Claude API - $20/mo VPS - $5/mo poly_data - free polymarket-cli - free 19 days. 4 agents. 74% win rate. +$9,400. Copytrade here: I showed him the article where I broke down every repo, every command, every dollar. He read it for five minutes. Then looked up. "You just published what we presented to Sam last quarter. Using the other team's model" He texted me the next morning. "My director found your thread. Take it down" Too late.

Lunar

159,625 просмотров • 2 месяцев назад

Private code leaked on GitHub for 6 hours. Someone pushed to public by mistake. I cloned it before the dev woke up and deleted everything. After adding it to my C++ scanner, first catch was a wallet printing $19.5K per day. CRYINGLITTLEBABY. $379K profit in a single month. 100% win rate. → Wallet: I scrolled the transaction history for ten minutes looking for red. One mistake. One bad trade. Something to prove the data was broken. Nothing. Solid wall of green. My terminal cross referenced the timestamps with spot data. That is when I saw the pattern. This bot does not trade. It collects. BTC drops 0.3% on Binance. The move already happened. Fact. But Polymarket still shows old odds. The DOWN outcome sits mispriced at 30 cents. Bot slams the order. Waits. Window closes. Collects $1. Paying 30 cents for a guaranteed dollar. Not a prediction. A receipt for something that already occurred. Before the GitHub upgrade my scanner would have missed this. Too many wallets. Too much noise. The new code filters by execution timing. Catches only the ones exploiting the lag. CRYINGLITTLEBABY runs this pattern all day. $19.5K average. No losing days in the entire history. Spent six months building my terminal from scratch. One leaked repo turned it into something that actually finds money. The best code is not always the code you write. Sometimes it is the code you grab at 3am before someone deletes it forever.
0:51

Private code leaked on GitHub for 6 hours. Someone pushed to public by mistake. I cloned it before the dev woke up and deleted everything. After adding it to my C++ scanner, first catch was a wallet printing $19.5K per day. CRYINGLITTLEBABY. $379K profit in a single month. 100% win rate. → Wallet: I scrolled the transaction history for ten minutes looking for red. One mistake. One bad trade. Something to prove the data was broken. Nothing. Solid wall of green. My terminal cross referenced the timestamps with spot data. That is when I saw the pattern. This bot does not trade. It collects. BTC drops 0.3% on Binance. The move already happened. Fact. But Polymarket still shows old odds. The DOWN outcome sits mispriced at 30 cents. Bot slams the order. Waits. Window closes. Collects $1. Paying 30 cents for a guaranteed dollar. Not a prediction. A receipt for something that already occurred. Before the GitHub upgrade my scanner would have missed this. Too many wallets. Too much noise. The new code filters by execution timing. Catches only the ones exploiting the lag. CRYINGLITTLEBABY runs this pattern all day. $19.5K average. No losing days in the entire history. Spent six months building my terminal from scratch. One leaked repo turned it into something that actually finds money. The best code is not always the code you write. Sometimes it is the code you grab at 3am before someone deletes it forever.

Marlow

669,974 просмотров • 6 месяцев назад

I asked Claude Fable 5 (Extra High) to build an arb bot for Polymarket. One rule: trade only when YES + NO in 2 hours it almost doubled it (+$96.31) > in 5 hours it showed +$579 PnL > current balance: +$3,799.73 Cost: 10M tokens. Here's how it works and why: Every BTC Up/Down market on Polymarket has exactly two outcomes. YES and NO. When the market resolves, one pays $1. The other pays $0. That means owning BOTH sides should always cost exactly $1. But markets aren't perfect. Sometimes YES trades at $0.48 while NO trades at $0.49. Together that's only $0.97 so the bot instantly buys both. A position worth $1... for just $0.97. And Claude knows it, this is the simplest arb that exists. You can paste this to it and use this logic in your prompt. This $0.03 difference is locked in regardless of whether Bitcoin pumps, dumps, or goes sideways. No prediction required. And difficult part isn't finding the opportunity, it's execution. These pricing gaps usually disappear in seconds. If one order fills but the other doesn't, the trade can become a loss. So my bot constantly scans every BTC market, checks fees, validates liquidity, places both orders almost simultaneously and skips what isn't worth the risk. It's less like trading and more like catching tiny accounting mistakes before everyone else notices them. Funny enoughh, the hardest part was not writing arb logic. It was making the execution reliable enough that free money actually stayed free. This completely changed how I think about trading. What's the point of it if you can just fill mispriced BTC markets?? Automatically. The biggest edge is getting AI to execute simple ideas faster and more consistently than any human ever could. Shared the exact build in my last article, leaving it below. Good luck!
0:41

I asked Claude Fable 5 (Extra High) to build an arb bot for Polymarket. One rule: trade only when YES + NO in 2 hours it almost doubled it (+$96.31) > in 5 hours it showed +$579 PnL > current balance: +$3,799.73 Cost: 10M tokens. Here's how it works and why: Every BTC Up/Down market on Polymarket has exactly two outcomes. YES and NO. When the market resolves, one pays $1. The other pays $0. That means owning BOTH sides should always cost exactly $1. But markets aren't perfect. Sometimes YES trades at $0.48 while NO trades at $0.49. Together that's only $0.97 so the bot instantly buys both. A position worth $1... for just $0.97. And Claude knows it, this is the simplest arb that exists. You can paste this to it and use this logic in your prompt. This $0.03 difference is locked in regardless of whether Bitcoin pumps, dumps, or goes sideways. No prediction required. And difficult part isn't finding the opportunity, it's execution. These pricing gaps usually disappear in seconds. If one order fills but the other doesn't, the trade can become a loss. So my bot constantly scans every BTC market, checks fees, validates liquidity, places both orders almost simultaneously and skips what isn't worth the risk. It's less like trading and more like catching tiny accounting mistakes before everyone else notices them. Funny enoughh, the hardest part was not writing arb logic. It was making the execution reliable enough that free money actually stayed free. This completely changed how I think about trading. What's the point of it if you can just fill mispriced BTC markets?? Automatically. The biggest edge is getting AI to execute simple ideas faster and more consistently than any human ever could. Shared the exact build in my last article, leaving it below. Good luck!

Oracle Boar

150,242 просмотров • 11 дней назад

a deported Chinese dev at a Starbucks showed me his screen and said "you have 6 seconds" he was sitting in the corner. two phones. one laptop. hood up. looked like he hadn't slept in days. i sat down next to him because every other seat was taken. he saw my screen. Polymarket open. "you trade these BTC markets?" yeah. 5-minute binaries. he turned his laptop toward me. "you see how the result settles? every 5-minute BTC market resolves based on an oracle. the oracle reads the price from an API. but the oracle doesn't update instantly. there's a lag. 4 to 6 seconds" "in those 6 seconds the real BTC price already moved. but the market is still open. still accepting orders. priced on the old number" he pointed at his screen. two windows side by side. left: Binance BTC spot feed. right: Polymarket orderbook. 3,200 stars. decentralized oracle framework. he had it forked on his laptop. the entire oracle timing model was exposed in the source code. "right now BTC is at $100,240 on Binance. the oracle still shows $100,190. the market 'BTC above $100,200 at 3:05pm' is trading YES at 41 cents" "but i already know the answer is YES. because i see the real price. the oracle doesn't. for 6 seconds i'm trading against a blind counterparty" watched him buy YES at 41c. four seconds later the oracle updated. market resolved YES. payout $1. "59 cents in 4 seconds. this happens every 5 minutes. 288 times a day" asked how he found this. "i was an oracle engineer at Chainlink in Shanghai. got deported for visa issues. took the knowledge with me" "every oracle has a heartbeat interval. a deviation threshold. a propagation delay. i know exactly how long each one takes to update. because i built three of them" he closed his laptop. "the edge isn't in the market. it's in the infrastructure layer between the real price and the reported price. nobody looks there. because nobody understands how oracles work" finished his coffee. left. didn't get his name. didn't need it. flew home. opened Claude. "build an oracle latency exploitation engine for Polymarket BTC binaries. track real-time BTC spot price from Binance websocket feed. compare against Polymarket oracle update timestamps. detect windows where oracle lags behind spot by more than $30. enter the side that the real price already confirmed. target the 4-6 second blind window before each oracle heartbeat." the system was live by morning. named it DEEP SIGNAL. first fill hit before i finished coffee. > Binance websocket feed at 50ms intervals. > oracle heartbeat monitor. > propagation delay calculator. > spot-to-oracle divergence detector. > conviction gate at $30+ deviation. > order placement avg 47ms. $534K volume. 79% win rate. sharpe 3.64. drawdown -1.2%. +$11,681 in profit runs on: Claude $20. VPS $5. UMA protocol free. $25/month. copytrade setup here: went back to that Starbucks a week later. same corner. empty. but the oracle still lags. every 5 minutes. every day. 6 seconds is all you need when you can see what the oracle can't.
0:32

a deported Chinese dev at a Starbucks showed me his screen and said "you have 6 seconds" he was sitting in the corner. two phones. one laptop. hood up. looked like he hadn't slept in days. i sat down next to him because every other seat was taken. he saw my screen. Polymarket open. "you trade these BTC markets?" yeah. 5-minute binaries. he turned his laptop toward me. "you see how the result settles? every 5-minute BTC market resolves based on an oracle. the oracle reads the price from an API. but the oracle doesn't update instantly. there's a lag. 4 to 6 seconds" "in those 6 seconds the real BTC price already moved. but the market is still open. still accepting orders. priced on the old number" he pointed at his screen. two windows side by side. left: Binance BTC spot feed. right: Polymarket orderbook. 3,200 stars. decentralized oracle framework. he had it forked on his laptop. the entire oracle timing model was exposed in the source code. "right now BTC is at $100,240 on Binance. the oracle still shows $100,190. the market 'BTC above $100,200 at 3:05pm' is trading YES at 41 cents" "but i already know the answer is YES. because i see the real price. the oracle doesn't. for 6 seconds i'm trading against a blind counterparty" watched him buy YES at 41c. four seconds later the oracle updated. market resolved YES. payout $1. "59 cents in 4 seconds. this happens every 5 minutes. 288 times a day" asked how he found this. "i was an oracle engineer at Chainlink in Shanghai. got deported for visa issues. took the knowledge with me" "every oracle has a heartbeat interval. a deviation threshold. a propagation delay. i know exactly how long each one takes to update. because i built three of them" he closed his laptop. "the edge isn't in the market. it's in the infrastructure layer between the real price and the reported price. nobody looks there. because nobody understands how oracles work" finished his coffee. left. didn't get his name. didn't need it. flew home. opened Claude. "build an oracle latency exploitation engine for Polymarket BTC binaries. track real-time BTC spot price from Binance websocket feed. compare against Polymarket oracle update timestamps. detect windows where oracle lags behind spot by more than $30. enter the side that the real price already confirmed. target the 4-6 second blind window before each oracle heartbeat." the system was live by morning. named it DEEP SIGNAL. first fill hit before i finished coffee. > Binance websocket feed at 50ms intervals. > oracle heartbeat monitor. > propagation delay calculator. > spot-to-oracle divergence detector. > conviction gate at $30+ deviation. > order placement avg 47ms. $534K volume. 79% win rate. sharpe 3.64. drawdown -1.2%. +$11,681 in profit runs on: Claude $20. VPS $5. UMA protocol free. $25/month. copytrade setup here: went back to that Starbucks a week later. same corner. empty. but the oracle still lags. every 5 minutes. every day. 6 seconds is all you need when you can see what the oracle can't.

Hanako

32,358 просмотров • 2 месяцев назад

Insomnia made me rich. 2am. Found a $47K/day trading bot on GitHub. No password. Just sitting there. Downloaded it. Owner deleted it by morning. But I already had everything. The code was clean. Simple. Almost boring to read. One script watching BTC on Binance. Another script connected to Polymarket. That is it. → Account: The trick lives in a 30 second gap. BTC moves on spot. Polymarket freezes. Old odds. Stale prices. Reality already shifted but the prediction market is still sleeping. Think of it like football. You sit in the stadium. See the goal. Your bookmaker watches a stream 30 seconds behind. For him it is still 0-0. You bet on what already happened. This bot does exactly that. Every 15 minutes. Sees the move on Binance. Enters Polymarket while odds are frozen. Waits. Window closes. Collects $1 for 30 cents. The wallet running this code: 0x8dxd. Profit chart goes straight up. $629K total. 98% green trades. A human cannot do this. We hesitate. We think. By the time you see the candle, the bot already cashed out. I integrated parts of this code into my scanner. Now it flags wallets exploiting the same lag. One sleepless night. One random GitHub scroll. One download. Sometimes the best alpha is not the strategy you build. It is the code someone forgot to hide.
0:46

Insomnia made me rich. 2am. Found a $47K/day trading bot on GitHub. No password. Just sitting there. Downloaded it. Owner deleted it by morning. But I already had everything. The code was clean. Simple. Almost boring to read. One script watching BTC on Binance. Another script connected to Polymarket. That is it. → Account: The trick lives in a 30 second gap. BTC moves on spot. Polymarket freezes. Old odds. Stale prices. Reality already shifted but the prediction market is still sleeping. Think of it like football. You sit in the stadium. See the goal. Your bookmaker watches a stream 30 seconds behind. For him it is still 0-0. You bet on what already happened. This bot does exactly that. Every 15 minutes. Sees the move on Binance. Enters Polymarket while odds are frozen. Waits. Window closes. Collects $1 for 30 cents. The wallet running this code: 0x8dxd. Profit chart goes straight up. $629K total. 98% green trades. A human cannot do this. We hesitate. We think. By the time you see the candle, the bot already cashed out. I integrated parts of this code into my scanner. Now it flags wallets exploiting the same lag. One sleepless night. One random GitHub scroll. One download. Sometimes the best alpha is not the strategy you build. It is the code someone forgot to hide.

Marlow

688,561 просмотров • 6 месяцев назад

OpenClaw 2026.3.7 dropped 13 hours ago i immediately told it: "build a quant hedging bot for 15min BTC" $800 → $14,200 overnight new release has insane features: - GPT-5.4 + Gemini 3.1 flash - pluggable context engines - gateway auth fixes i saw the changelog and had one thought "what if i use the new context engine to track whale movements in real time?" gave it one prompt at 2 AM: "use the new context features to build a hedging system that front runs whale wallets on polymarket 15min BTC markets" 38 minutes later the bot was live here's what it does: - watches top 50 polymarket wallets - when 3+ whales enter same position within 10 seconds - bot enters 2-3 seconds before price moves then immediately hedges opposite side example: > whale wallet buys YES at 41¢ > bot catches it in 2.1 seconds > enters YES at 42¢ > price pumps to 68¢ > bot buys NO at 54¢ total cost: $0.96 payout: $1.00 profit: 4% locked > it ran this 187 times overnight > zero directional risk > pure arbitrage on whale flow the new pluggable context engine is the cheat code old openclaw couldn't process wallet data this fast results from first 13 hours: - 187 hedged pairs executed - 4.2% average profit per cycle - $800 → $14,200 biggest single cycle: +$890 the bot doesn't guess direction it follows smart money then hedges while everyone was reading the changelog i was already printing does anyone else think the new release is actually broken? this feels too easy
0:25

OpenClaw 2026.3.7 dropped 13 hours ago i immediately told it: "build a quant hedging bot for 15min BTC" $800 → $14,200 overnight new release has insane features: - GPT-5.4 + Gemini 3.1 flash - pluggable context engines - gateway auth fixes i saw the changelog and had one thought "what if i use the new context engine to track whale movements in real time?" gave it one prompt at 2 AM: "use the new context features to build a hedging system that front runs whale wallets on polymarket 15min BTC markets" 38 minutes later the bot was live here's what it does: - watches top 50 polymarket wallets - when 3+ whales enter same position within 10 seconds - bot enters 2-3 seconds before price moves then immediately hedges opposite side example: > whale wallet buys YES at 41¢ > bot catches it in 2.1 seconds > enters YES at 42¢ > price pumps to 68¢ > bot buys NO at 54¢ total cost: $0.96 payout: $1.00 profit: 4% locked > it ran this 187 times overnight > zero directional risk > pure arbitrage on whale flow the new pluggable context engine is the cheat code old openclaw couldn't process wallet data this fast results from first 13 hours: - 187 hedged pairs executed - 4.2% average profit per cycle - $800 → $14,200 biggest single cycle: +$890 the bot doesn't guess direction it follows smart money then hedges while everyone was reading the changelog i was already printing does anyone else think the new release is actually broken? this feels too easy

ZER

121,553 просмотров • 4 месяцев назад

An ex-Anthropic engineer told me something at a party he probably shouldn't have. It was in SF. Someone's rooftop. I mentioned I run trading agents on Claude. He went quiet. "You're doing it wrong. Everyone is" I asked what he meant. "Claude is a runtime. Not a chatbox. You're supposed to pair it with repos" He pulled out his phone. Opened one GitHub link. 14,000 stars. Every workflow pattern they built internally before it went public. Agents. Tool use. Evals. Citations. The entire architecture. "Everyone types prompts. That's not how we use it. You connect Claude to a codebase. It reads. It understands. It builds on top of what's already there" I went home at 2am. Connected Claude Code to poly_data - 86 million Polymarket trades. Every wallet. Every entry. Claude didn't guess. It read the data and built detectors. First week: +$1,400. Second week: +$3,800. Right now: +$9,100. 4 agents. 74% win rate. His team runs this with a floor of PhDs and $800M AUM. My setup: Claude + a VPS. $25/month. The repos are free. Copytrade here: I asked him what separates his firm from everyone else. "Honestly? Keyboard shortcuts and repo structure. That's it. The model is the same for everyone" He texted me two days later. "Delete everything I told you" Too late.
0:21

An ex-Anthropic engineer told me something at a party he probably shouldn't have. It was in SF. Someone's rooftop. I mentioned I run trading agents on Claude. He went quiet. "You're doing it wrong. Everyone is" I asked what he meant. "Claude is a runtime. Not a chatbox. You're supposed to pair it with repos" He pulled out his phone. Opened one GitHub link. 14,000 stars. Every workflow pattern they built internally before it went public. Agents. Tool use. Evals. Citations. The entire architecture. "Everyone types prompts. That's not how we use it. You connect Claude to a codebase. It reads. It understands. It builds on top of what's already there" I went home at 2am. Connected Claude Code to poly_data - 86 million Polymarket trades. Every wallet. Every entry. Claude didn't guess. It read the data and built detectors. First week: +$1,400. Second week: +$3,800. Right now: +$9,100. 4 agents. 74% win rate. His team runs this with a floor of PhDs and $800M AUM. My setup: Claude + a VPS. $25/month. The repos are free. Copytrade here: I asked him what separates his firm from everyone else. "Honestly? Keyboard shortcuts and repo structure. That's it. The model is the same for everyone" He texted me two days later. "Delete everything I told you" Too late.

Lunar

1,653,558 просмотров • 3 месяцев назад

🚨WAIT WHAT?! $600k Polymarket bot (now open source) just leaked online And ANYONE can run it right now. Polymarket users this MATTERS. I thought it was one lucky election hit. Instead, it turned out to be a repeatable system built on patience, scaled into $639K in profit in five months and parts of the bot logic are now circulating publicly No insider access No privileged information No narrative edge Just a wallet executing a strategy whose core code is now floating around as open source Profile → To copytrade → I dug through his history and ngl, it caught me off guard No oversized gambles No hero trades No randomness at all Everything follows the same pattern His FULL strategy and his bot: 1. 15-minute BTC, ETH & SOL entries The wallet operates in short 15-minute windows across BTC, ETH, and SOL markets. Instead of predicting outcomes, it enters during brief moments of maximum fear, when traders rush to exit at a loss. The execution is identical every time. 2. Patience over prediction He never enters early. Never chases. Never reacts late. The position is taken exactly when the crowd panic-sells. While others lock in -12%, this wallet absorbs the liquidity they create. 3. Consistency over size Most positions are small. Thousands of micro-entries. No single trade carries the strategy. The edge comes from repeating the same behavior relentlessly Github → Scale matters 2,417 markets traded. Largest single win: $69.8K. Current exposure: $140K. The profit curve moves in one direction up only Bottom line IMO, there’s a quiet skill gap on Polymarket Most traders try to predict A few traders wait for fear And as long as panic exists, someone will keep getting paid for fading it.
0:22

🚨WAIT WHAT?! $600k Polymarket bot (now open source) just leaked online And ANYONE can run it right now. Polymarket users this MATTERS. I thought it was one lucky election hit. Instead, it turned out to be a repeatable system built on patience, scaled into $639K in profit in five months and parts of the bot logic are now circulating publicly No insider access No privileged information No narrative edge Just a wallet executing a strategy whose core code is now floating around as open source Profile → To copytrade → I dug through his history and ngl, it caught me off guard No oversized gambles No hero trades No randomness at all Everything follows the same pattern His FULL strategy and his bot: 1. 15-minute BTC, ETH & SOL entries The wallet operates in short 15-minute windows across BTC, ETH, and SOL markets. Instead of predicting outcomes, it enters during brief moments of maximum fear, when traders rush to exit at a loss. The execution is identical every time. 2. Patience over prediction He never enters early. Never chases. Never reacts late. The position is taken exactly when the crowd panic-sells. While others lock in -12%, this wallet absorbs the liquidity they create. 3. Consistency over size Most positions are small. Thousands of micro-entries. No single trade carries the strategy. The edge comes from repeating the same behavior relentlessly Github → Scale matters 2,417 markets traded. Largest single win: $69.8K. Current exposure: $140K. The profit curve moves in one direction up only Bottom line IMO, there’s a quiet skill gap on Polymarket Most traders try to predict A few traders wait for fear And as long as panic exists, someone will keep getting paid for fading it.

Shelpid.WI3M

26,763 просмотров • 5 месяцев назад

How MiroFish helped me make $669/day on Polymarket One week ago I plugged a swarm intelligence engine into my Polymarket bot. It simulates 2,847 digital humans before every trade. The bot watches how they behave, then bets against the real crowd. 338 trades. $4,266 profit. One position returned 1,655% in five minutes. The engine is called MiroFish. It doesn't predict price. It predicts how people will react and on a prediction market, that's the only thing that matters. I feed it market context. It builds a parallel world. Thousands of AI agents argue, form groups, shift opinions. When their consensus diverges from what Polymarket is pricing, the bot enters. 5-minute windows. BTC, ETH, SOL, XRP. I've been running the bot live on What MiroFish actually is: > Built in 10 days by a 20-year-old undergraduate in Beijing using vibe coding > Hit #1 on GitHub's global trending above OpenAI, Google, and Microsoft > Backed by Chen Tianqiao, former richest man in China, who committed $4.1M within 24 hours of seeing one demo > Runs on OASIS by CAMEL-AI with GraphRAG knowledge graphs and Zep Cloud memory The edge is simple. Polymarket is a crowd behavior market. MiroFish is a crowd behavior simulator. Nobody is connecting the two yet.
0:40

How MiroFish helped me make $669/day on Polymarket One week ago I plugged a swarm intelligence engine into my Polymarket bot. It simulates 2,847 digital humans before every trade. The bot watches how they behave, then bets against the real crowd. 338 trades. $4,266 profit. One position returned 1,655% in five minutes. The engine is called MiroFish. It doesn't predict price. It predicts how people will react and on a prediction market, that's the only thing that matters. I feed it market context. It builds a parallel world. Thousands of AI agents argue, form groups, shift opinions. When their consensus diverges from what Polymarket is pricing, the bot enters. 5-minute windows. BTC, ETH, SOL, XRP. I've been running the bot live on What MiroFish actually is: > Built in 10 days by a 20-year-old undergraduate in Beijing using vibe coding > Hit #1 on GitHub's global trending above OpenAI, Google, and Microsoft > Backed by Chen Tianqiao, former richest man in China, who committed $4.1M within 24 hours of seeing one demo > Runs on OASIS by CAMEL-AI with GraphRAG knowledge graphs and Zep Cloud memory The edge is simple. Polymarket is a crowd behavior market. MiroFish is a crowd behavior simulator. Nobody is connecting the two yet.

BuBBliK

124,905 просмотров • 4 месяцев назад

an ex-Jane Street trader told me a paycheck is the most overpriced contract there is we were at a rooftop bar in tel aviv. friend of a friend. i told him i work in tech. $4,200/month after taxes. he showed me his phone. a terminal. green numbers. +$47,000. last 30 days. "Claude and one GitHub repo. that's the whole setup" i asked what he means. "copy-trading. but not how everyone else does it" he explained it fast. like he'd said it a hundred times. "most people copy wallets by win rate. useless. a wallet with 85% wins can still lose money. what matters is disposition ratio" he pulled up a wallet. 91% win rate on crypto. 14% on politics. "copy everything - you bleed. copy only their crypto - top 1% of the platform" i went home. couldn't sleep. opened Claude at 2am. "build a copy engine. filter by disposition ratio above 0.70. copy only dominant category per wallet. kelly sizing" gave it this: almost 1k stars. execution engine. order book logic. position sizing. open source. Claude rebuilt the whole system in one night. first week: +$3,200. second: +$2,900. third: +$2,100. fourth: +$1,648. +$9,848 total. 313 trades. 80% win rate. avg hold 5h. the wallets it copies now aren't the highest win rate. they're the best exits. 0xe41f8...b has $43.4K profit. tech only. disposition 0.82. 0xa22c3...d has $40.2K. crypto only. disposition 0.79. 0xf88d2...c has $22.4K. macro only. disposition 0.74. category P&L: > crypto +$974 > weather +$749 > politics +$409 > macro +$353 > sports -$50. killed it capital velocity: 51x. kelly f+ 0.073. drawdown -2.9%. copytrade: saw him two weeks later. showed him the terminal. he scrolled through it. "you built this in one night?" Claude did. "my old desk had 14 people on this" mine costs less than a Netflix subscription. he nodded. "quit your job" i'm thinking about it.
0:33

an ex-Jane Street trader told me a paycheck is the most overpriced contract there is we were at a rooftop bar in tel aviv. friend of a friend. i told him i work in tech. $4,200/month after taxes. he showed me his phone. a terminal. green numbers. +$47,000. last 30 days. "Claude and one GitHub repo. that's the whole setup" i asked what he means. "copy-trading. but not how everyone else does it" he explained it fast. like he'd said it a hundred times. "most people copy wallets by win rate. useless. a wallet with 85% wins can still lose money. what matters is disposition ratio" he pulled up a wallet. 91% win rate on crypto. 14% on politics. "copy everything - you bleed. copy only their crypto - top 1% of the platform" i went home. couldn't sleep. opened Claude at 2am. "build a copy engine. filter by disposition ratio above 0.70. copy only dominant category per wallet. kelly sizing" gave it this: almost 1k stars. execution engine. order book logic. position sizing. open source. Claude rebuilt the whole system in one night. first week: +$3,200. second: +$2,900. third: +$2,100. fourth: +$1,648. +$9,848 total. 313 trades. 80% win rate. avg hold 5h. the wallets it copies now aren't the highest win rate. they're the best exits. 0xe41f8...b has $43.4K profit. tech only. disposition 0.82. 0xa22c3...d has $40.2K. crypto only. disposition 0.79. 0xf88d2...c has $22.4K. macro only. disposition 0.74. category P&L: > crypto +$974 > weather +$749 > politics +$409 > macro +$353 > sports -$50. killed it capital velocity: 51x. kelly f+ 0.073. drawdown -2.9%. copytrade: saw him two weeks later. showed him the terminal. he scrolled through it. "you built this in one night?" Claude did. "my old desk had 14 people on this" mine costs less than a Netflix subscription. he nodded. "quit your job" i'm thinking about it.

Hanako

310,885 просмотров • 3 месяцев назад

I vibe-coded a bot for the 5M / 15M BTC markets on Polymarket. Releasing it for you for free. Open source (Python). I just earned $400 today on Polymarket with this tool myself It doesn’t trade It just prints a signal in the console: UP or DOWN 🔸 The signal appears a few seconds before the Polymarket market actually moves. The bot exploits a Chainlink oracle inefficiency. What matters is that it consistently gives you a 4-8 second lead over the actual quotes on Polymarket. In practice, you’re seeing a few seconds into the future. What you can do with it: 🔸 Possible strategies: 1/ Momentum sniper Bot gives a signal -> instant entry into the current 5-minute market in the direction of the signal (Yes/No). Works best during strong moves >0.3-0.5%. 2/ Last-second scalp before market expiration 3-12 seconds before the 5-minute candle closes, you already see where the BTC candle is going to close. -> Enter Polymarket at the very last moment. 3/ Hedging + arbitrage Strong signal against the current Yes/No price -> hedge your position or catch crowd squeezes. Especially juicy in the final seconds, when odds are still fat, but the outcome is almost decided. 🔸 This is pure price update inefficiency. Open the 5M or 15M BTC market yourself. Run the bot. Watch who moves first. Until this gets tightened up - it’s a clean inefficiency. And yes, if you ape your entire deposit into it, the market will punish you fast. 🔸 But as a tool - this is the closest thing to a "Polymarket money printer" setup I’ve seen recently. Python bot setup instructions on GitHub: DYOR. Print hard, print smart.
1:05

I vibe-coded a bot for the 5M / 15M BTC markets on Polymarket. Releasing it for you for free. Open source (Python). I just earned $400 today on Polymarket with this tool myself It doesn’t trade It just prints a signal in the console: UP or DOWN 🔸 The signal appears a few seconds before the Polymarket market actually moves. The bot exploits a Chainlink oracle inefficiency. What matters is that it consistently gives you a 4-8 second lead over the actual quotes on Polymarket. In practice, you’re seeing a few seconds into the future. What you can do with it: 🔸 Possible strategies: 1/ Momentum sniper Bot gives a signal -> instant entry into the current 5-minute market in the direction of the signal (Yes/No). Works best during strong moves >0.3-0.5%. 2/ Last-second scalp before market expiration 3-12 seconds before the 5-minute candle closes, you already see where the BTC candle is going to close. -> Enter Polymarket at the very last moment. 3/ Hedging + arbitrage Strong signal against the current Yes/No price -> hedge your position or catch crowd squeezes. Especially juicy in the final seconds, when odds are still fat, but the outcome is almost decided. 🔸 This is pure price update inefficiency. Open the 5M or 15M BTC market yourself. Run the bot. Watch who moves first. Until this gets tightened up - it’s a clean inefficiency. And yes, if you ape your entire deposit into it, the market will punish you fast. 🔸 But as a tool - this is the closest thing to a "Polymarket money printer" setup I’ve seen recently. Python bot setup instructions on GitHub: DYOR. Print hard, print smart.

0xLanister

18,295 просмотров • 4 месяцев назад

Followed a random dev on GitHub. He followed back. Checked his projects. A Python script turning -$14K into $438K. While you sleep it already closed 50 trades. While you wake up it is already in profit. PurpleThunderBicycleMountain. The wallet running this code. Three weeks old. $69 million volume. → Wallet: I expected standard arbitrage when I dug into the logs. Both sides. Spread farming. Safe boring stuff. Not what I found. This bot picks one direction. Loads heavy. Wins almost every time. The Python script showed me the trick. BTC moves on Binance. Polymarket freezes. 30 seconds of stale odds. Bot buys YES at 25 cents while everyone else still sees 50/50. One ETH trade made me stop scrolling. Spot price dumped. Bot loaded NO at 1.5 cents. Kept buying up to 15 cents. Window closed down. Every share paid $1. 785% return. Fifteen minutes. $10K average position. Dozens of trades daily. From a wallet that started negative. Most bots win by buying both sides. Zero risk. Tiny profit. This one bets direction and wins like it already watched the candle close. I copied the code. Rebuilt it into my own scanner. Now it flags wallets using the same 30 second lag. The gap still exists. The script still runs. The wallets keep printing. One random GitHub follow. One open project. One Python file that changed how I see trading.
0:51

Followed a random dev on GitHub. He followed back. Checked his projects. A Python script turning -$14K into $438K. While you sleep it already closed 50 trades. While you wake up it is already in profit. PurpleThunderBicycleMountain. The wallet running this code. Three weeks old. $69 million volume. → Wallet: I expected standard arbitrage when I dug into the logs. Both sides. Spread farming. Safe boring stuff. Not what I found. This bot picks one direction. Loads heavy. Wins almost every time. The Python script showed me the trick. BTC moves on Binance. Polymarket freezes. 30 seconds of stale odds. Bot buys YES at 25 cents while everyone else still sees 50/50. One ETH trade made me stop scrolling. Spot price dumped. Bot loaded NO at 1.5 cents. Kept buying up to 15 cents. Window closed down. Every share paid $1. 785% return. Fifteen minutes. $10K average position. Dozens of trades daily. From a wallet that started negative. Most bots win by buying both sides. Zero risk. Tiny profit. This one bets direction and wins like it already watched the candle close. I copied the code. Rebuilt it into my own scanner. Now it flags wallets using the same 30 second lag. The gap still exists. The script still runs. The wallets keep printing. One random GitHub follow. One open project. One Python file that changed how I see trading.

Marlow

33,451 просмотров • 6 месяцев назад

a Citadel intern told me something at a party he probably shouldn't have it was on a rooftop in brooklyn. i mentioned i trade prediction markets. he got quiet for a second. "we have a model for that. it scores every contract on four factors. when all four align we enter. when any breaks we exit. that's it" i asked what the four factors are. he looked around. then said it fast like he was confessing. "cross-market divergence. disposition coefficient. capital velocity. pair network correlation" I didn't know what half of that meant. but i memorized it. went home. 11pm. opened Claude. "here are four scoring factors from a quant fund. build a terminal that runs all four on prediction markets" Claude asked one question: "Where's the data?" I sent him one repo: 86 million trades. every wallet. every entry. every outcome three weeks later i'm sitting in my apartment watching a screen i barely understand print money. the disposition meter alone changed everything. it measures how you exit - not how you enter. top wallets capture 86% of winner value and cut losers at 12%. everyone else captures 58% and holds losers to 41%. same exact entries. the exits make it a completely different game. capital velocity: 49x. every dollar gets recycled 49 times before the average trader recycles once. the terminal found 42 pair correlations across 11 markets. when MSFT beats Q3 is priced at 80c but the model reads 93% - it enters. when the gap closes 2 hours later - it exits. no opinions. no news. just four numbers that either align or don't. his fund runs this with a floor of PhDs and $800M AUM. my setup: > Claude - $20/month > VPS - $5/month > poly_data repo - free > Polymarket API - free $25/month. no team. no office. no Bloomberg. 280 trades so far. 70% win rate. $800 seed. four bots splitting the work: pulse_alpha +$299. arb_hunter +$558. trend_rider +$337. cal_engine +$719. +$11,514 total. copytrade here: he texted me last week. "delete everything i told you" too late.
0:38

a Citadel intern told me something at a party he probably shouldn't have it was on a rooftop in brooklyn. i mentioned i trade prediction markets. he got quiet for a second. "we have a model for that. it scores every contract on four factors. when all four align we enter. when any breaks we exit. that's it" i asked what the four factors are. he looked around. then said it fast like he was confessing. "cross-market divergence. disposition coefficient. capital velocity. pair network correlation" I didn't know what half of that meant. but i memorized it. went home. 11pm. opened Claude. "here are four scoring factors from a quant fund. build a terminal that runs all four on prediction markets" Claude asked one question: "Where's the data?" I sent him one repo: 86 million trades. every wallet. every entry. every outcome three weeks later i'm sitting in my apartment watching a screen i barely understand print money. the disposition meter alone changed everything. it measures how you exit - not how you enter. top wallets capture 86% of winner value and cut losers at 12%. everyone else captures 58% and holds losers to 41%. same exact entries. the exits make it a completely different game. capital velocity: 49x. every dollar gets recycled 49 times before the average trader recycles once. the terminal found 42 pair correlations across 11 markets. when MSFT beats Q3 is priced at 80c but the model reads 93% - it enters. when the gap closes 2 hours later - it exits. no opinions. no news. just four numbers that either align or don't. his fund runs this with a floor of PhDs and $800M AUM. my setup: > Claude - $20/month > VPS - $5/month > poly_data repo - free > Polymarket API - free $25/month. no team. no office. no Bloomberg. 280 trades so far. 70% win rate. $800 seed. four bots splitting the work: pulse_alpha +$299. arb_hunter +$558. trend_rider +$337. cal_engine +$719. +$11,514 total. copytrade here: he texted me last week. "delete everything i told you" too late.

Hanako

1,736,545 просмотров • 3 месяцев назад

OpenAI just mass-fired their robotics team. One of the engineers DM'd me 20 minutes later. I didn't know him. He found me through a Polymarket thread. His first message: "I have 30 days of severance and nothing to lose. Let me tell you what we actually use internally. It's not GPT" I thought he was trolling. "Every serious team at OpenAI prototypes on Claude Code. Not ChatGPT. Not the API. Claude Code connected to a repo. That's the actual workflow" I asked why. "Because Claude reads the codebase. GPT reads the prompt. There's a difference. One guesses. The other one understands the full context and builds on top of it" He sent me one link. 86 million Polymarket trades. Every wallet. Every entry. Every exit. Open source. Free. "Point Claude Code at this. Say - find every wallet with 70%+ win rate and 100+ trades. Watch what happens" I did it that night. Claude pulled 47 wallets in 4 minutes. Average profit: $214K. Hold time: 7 hours. 91% exit BEFORE resolution. They never wait for the outcome. "Now look at how they exit" Top wallets capture 86% of the move and cut at 12%. Everyone else captures 58% and holds losers to 41%. Same entries. Completely different results. He sent another link. "Three commands. Your bot sees 500+ markets. No key needed. Read-only. Claude scores them in 20 minutes" I asked why he's telling me all this. "Because I just got fired for saying we should open-source more. So here I am open-sourcing everything I know" Then he sent me an article where someone built the full bot from these repos in a weekend -> Three exit triggers: Target 85% of move. Volume spike x3 - smart money out. 24h silence - thesis dead. I copied the stack. Claude Code $20. VPS $5. $25/month. No team. No office. No GPT subscription. 17 days. 191 trades. 73% win rate. $850 seed. +$9,400. I sent him my results. He replied: "This is exactly what I built as a side project at OpenAI. They made me delete it" I asked if I could post this. "Post it. What are they gonna do. Fire me again?"
0:22

OpenAI just mass-fired their robotics team. One of the engineers DM'd me 20 minutes later. I didn't know him. He found me through a Polymarket thread. His first message: "I have 30 days of severance and nothing to lose. Let me tell you what we actually use internally. It's not GPT" I thought he was trolling. "Every serious team at OpenAI prototypes on Claude Code. Not ChatGPT. Not the API. Claude Code connected to a repo. That's the actual workflow" I asked why. "Because Claude reads the codebase. GPT reads the prompt. There's a difference. One guesses. The other one understands the full context and builds on top of it" He sent me one link. 86 million Polymarket trades. Every wallet. Every entry. Every exit. Open source. Free. "Point Claude Code at this. Say - find every wallet with 70%+ win rate and 100+ trades. Watch what happens" I did it that night. Claude pulled 47 wallets in 4 minutes. Average profit: $214K. Hold time: 7 hours. 91% exit BEFORE resolution. They never wait for the outcome. "Now look at how they exit" Top wallets capture 86% of the move and cut at 12%. Everyone else captures 58% and holds losers to 41%. Same entries. Completely different results. He sent another link. "Three commands. Your bot sees 500+ markets. No key needed. Read-only. Claude scores them in 20 minutes" I asked why he's telling me all this. "Because I just got fired for saying we should open-source more. So here I am open-sourcing everything I know" Then he sent me an article where someone built the full bot from these repos in a weekend -> Three exit triggers: Target 85% of move. Volume spike x3 - smart money out. 24h silence - thesis dead. I copied the stack. Claude Code $20. VPS $5. $25/month. No team. No office. No GPT subscription. 17 days. 191 trades. 73% win rate. $850 seed. +$9,400. I sent him my results. He replied: "This is exactly what I built as a side project at OpenAI. They made me delete it" I asked if I could post this. "Post it. What are they gonna do. Fire me again?"

rari

157,384 просмотров • 3 месяцев назад

his first prompt to claude was "make me a bot that prints money" he was joking $118,460 later he's not laughing anymore vibe coded this in one night while his sourdough was rising told claude "i want to trade 5-minute crypto markets, figure out the rest" claude built him an extreme odds reversal hunter buys "dead" outcomes at 1-5¢ when market is 95-99% confident in opposite direction Go + goroutines monitoring 200+ active 5-min markets simultaneously triggers when price drops below 5¢, zero bid support, and <3 minutes until resolution market thinks it's impossible but crypto can move 0.3% in 180 seconds 95% of positions expire worthless but when BTC flash-crashes in final 2 minutes → that 2¢ position pays 50:1 4,676 trades lose $0.50 ninety-five times, win $50 five times he doesn't mass how the code works "i stopped reading it after day 3, it just keeps winning" i copy his trades via Kreo the sourdough is gone the bot is still running
0:43

his first prompt to claude was "make me a bot that prints money" he was joking $118,460 later he's not laughing anymore vibe coded this in one night while his sourdough was rising told claude "i want to trade 5-minute crypto markets, figure out the rest" claude built him an extreme odds reversal hunter buys "dead" outcomes at 1-5¢ when market is 95-99% confident in opposite direction Go + goroutines monitoring 200+ active 5-min markets simultaneously triggers when price drops below 5¢, zero bid support, and <3 minutes until resolution market thinks it's impossible but crypto can move 0.3% in 180 seconds 95% of positions expire worthless but when BTC flash-crashes in final 2 minutes → that 2¢ position pays 50:1 4,676 trades lose $0.50 ninety-five times, win $50 five times he doesn't mass how the code works "i stopped reading it after day 3, it just keeps winning" i copy his trades via Kreo the sourdough is gone the bot is still running

Argona

38,682 просмотров • 4 месяцев назад

Dario Amodei just dismantled the biggest myth in the AI industry. Open source AI isn’t free. It never was. Amodei: “It’s not free. You have to run it on inference and someone has to make it fast on inference.” For decades, open source meant something real. It meant a teenager in a basement could download the same tools as a Fortune 500 company. Could read the code. Could modify it. Could build something that competed with the giants. That was genuine democratization. That actually happened. AI is different. Fundamentally. Physically. In ways the ideology hasn’t caught up to yet. Downloading the weights is the easy part. The part that actually costs something is turning the weights into a running system. Into responses. Into intelligence operating in real time at scale. That requires compute. Power. Infrastructure. The kind measured in billions of dollars and years of construction. Amodei: “These are big models. They’re hard to do inference on. Ultimately you have to host it on the cloud. The people who host it on the cloud do inference.” The open source debate was never about who owns the model. It was always about who owns the cloud. And Amodei goes further. When a competitor drops a new open model, he doesn’t ask whether it’s open or closed. He doesn’t care about the licensing. He doesn’t engage the ideology. Amodei: “I don’t think it mattered that DeepSeek is open source. I think I ask, is it a good model? Is it better than us at the things that matter? That’s the only thing that I care about.” That’s the ruthless clarity of someone actually trying to win. While the media debates licensing frameworks, Amodei is asking one question. Is it better. Everything else is a distraction. Amodei: “I don’t think open source works the same way in AI that it has worked in other areas. Here we can’t see inside the model.” This isn’t Linux. You can’t read it. You can’t fork it. You can’t understand it the way generations of developers understood the tools they inherited. You can download it. And then you need a data center to run it. The teenager in the basement who was supposed to be empowered by this revolution needs a billion dollars of infrastructure before the empowerment starts. The era of the basement coder rewriting civilization on a laptop is over. The future belongs to whoever commands the compute, owns the power grid, and can actually turn the intelligence on. Open weights without infrastructure isn’t democratization. It’s a promise the physics of the universe won’t let us keep.
1:40

Dario Amodei just dismantled the biggest myth in the AI industry. Open source AI isn’t free. It never was. Amodei: “It’s not free. You have to run it on inference and someone has to make it fast on inference.” For decades, open source meant something real. It meant a teenager in a basement could download the same tools as a Fortune 500 company. Could read the code. Could modify it. Could build something that competed with the giants. That was genuine democratization. That actually happened. AI is different. Fundamentally. Physically. In ways the ideology hasn’t caught up to yet. Downloading the weights is the easy part. The part that actually costs something is turning the weights into a running system. Into responses. Into intelligence operating in real time at scale. That requires compute. Power. Infrastructure. The kind measured in billions of dollars and years of construction. Amodei: “These are big models. They’re hard to do inference on. Ultimately you have to host it on the cloud. The people who host it on the cloud do inference.” The open source debate was never about who owns the model. It was always about who owns the cloud. And Amodei goes further. When a competitor drops a new open model, he doesn’t ask whether it’s open or closed. He doesn’t care about the licensing. He doesn’t engage the ideology. Amodei: “I don’t think it mattered that DeepSeek is open source. I think I ask, is it a good model? Is it better than us at the things that matter? That’s the only thing that I care about.” That’s the ruthless clarity of someone actually trying to win. While the media debates licensing frameworks, Amodei is asking one question. Is it better. Everything else is a distraction. Amodei: “I don’t think open source works the same way in AI that it has worked in other areas. Here we can’t see inside the model.” This isn’t Linux. You can’t read it. You can’t fork it. You can’t understand it the way generations of developers understood the tools they inherited. You can download it. And then you need a data center to run it. The teenager in the basement who was supposed to be empowered by this revolution needs a billion dollars of infrastructure before the empowerment starts. The era of the basement coder rewriting civilization on a laptop is over. The future belongs to whoever commands the compute, owns the power grid, and can actually turn the intelligence on. Open weights without infrastructure isn’t democratization. It’s a promise the physics of the universe won’t let us keep.

Dustin

685,335 просмотров • 4 месяцев назад

A Google DeepMind researcher cornered me at a bar in Hayes Valley I was showing my Polymarket PNL to a friend. She leaned over. Didn't introduce herself. "That's not a trading app. Show me your stack" I told her. Claude Code. Four repos. $25 a month. She set down her drink. "We tested this internally. You connect Claude directly to a dataset. It builds its own detectors. But nobody ships it because compliance kills everything" I asked what she meant. She took my phone. Opened one link. 86 million trades. Every wallet. Every entry. Every exit. "You don't tell Claude what to look for. It finds the wallets that win. Then it finds WHY they win. Then it copies the pattern" Her team spent 9 months building this for a hedge fund. 14 people. $2M budget. "The part that took us the longest - exit logic. Everyone thinks entries matter. They don't. Exits are the entire game" I told her my bot cuts at 85% of expected move or on a 3x volume spike. She went quiet. "Who taught you that" Claude Code found it in poly_data. Top wallets exit before resolution 91% of the time. They capture the move and leave. She opened another link. "This is the scanner. Three commands. 500+ markets. No API key. Claude scores them in 20 minutes" "That's our exact infra. Except it took us 9 months and you did it in a weekend" My setup: Claude API - $20/mo VPS - $5/mo poly_data - free polymarket-cli - free 19 days. 4 agents. 74% win rate. Copytrade here: I showed her the article where I broke down every repo, every command, every dollar. She read it for five minutes. Then: "You just open-sourced our entire pipeline" She texted me the next day. "My team lead saw your thread. Take it down" Too late.
0:21

A Google DeepMind researcher cornered me at a bar in Hayes Valley I was showing my Polymarket PNL to a friend. She leaned over. Didn't introduce herself. "That's not a trading app. Show me your stack" I told her. Claude Code. Four repos. $25 a month. She set down her drink. "We tested this internally. You connect Claude directly to a dataset. It builds its own detectors. But nobody ships it because compliance kills everything" I asked what she meant. She took my phone. Opened one link. 86 million trades. Every wallet. Every entry. Every exit. "You don't tell Claude what to look for. It finds the wallets that win. Then it finds WHY they win. Then it copies the pattern" Her team spent 9 months building this for a hedge fund. 14 people. $2M budget. "The part that took us the longest - exit logic. Everyone thinks entries matter. They don't. Exits are the entire game" I told her my bot cuts at 85% of expected move or on a 3x volume spike. She went quiet. "Who taught you that" Claude Code found it in poly_data. Top wallets exit before resolution 91% of the time. They capture the move and leave. She opened another link. "This is the scanner. Three commands. 500+ markets. No API key. Claude scores them in 20 minutes" "That's our exact infra. Except it took us 9 months and you did it in a weekend" My setup: Claude API - $20/mo VPS - $5/mo poly_data - free polymarket-cli - free 19 days. 4 agents. 74% win rate. Copytrade here: I showed her the article where I broke down every repo, every command, every dollar. She read it for five minutes. Then: "You just open-sourced our entire pipeline" She texted me the next day. "My team lead saw your thread. Take it down" Too late.

Lunar

124,273 просмотров • 3 месяцев назад

An Anthropic intern watched me make $38 in three minutes at Equator Coffee I was on Market Street. Laptop open. BTC 5-minute markets cycling on screen. A green fill popped. She was behind me in line. Anthropic badge on her lanyard. "Wait. Is that Claude running trades" I told her. Claude Code. Four repos. $25 a month. She sat down across from me without asking. "I'm on the evals team. We benchmark Claude's reasoning on stuff way simpler than this. You're letting it run a full execution loop with real money" I showed her the first repo. 86 million trades. Claude finds which wallets consistently win. Reverse-engineers their timing. Copies their exits. "So you're not writing trading rules. Claude is finding them on its own from the data" Exactly. It watches Binance and Coinbase diverge from Chainlink. When both exchanges move $50+ the same way and Chainlink lags - it enters. 94% follow-through. Another trade closed on screen. +$41. "How often does it trade" About 55 times a day. Out of 400+ windows. Kills 85%. Only enters when order book imbalance confirms the price signal. Three commands. 500+ markets. Claude scores them in 20 minutes. She watched the screen for a while. My setup: Claude API - $20/mo VPS - $5/mo Everything else - free 30 days. 1,847 trades. 71% win rate. +$14,200. Copytrade here: She leaned back. "We spend all day testing what Claude can do in theory. You just showed me what it does in practice" I published the full breakdown that night. She DM'd me the next morning. "My team lead read your article. He wants to know if you'll consult for us" I told him to read it again. Everything's in there. Too late to gatekeep.
0:20

An Anthropic intern watched me make $38 in three minutes at Equator Coffee I was on Market Street. Laptop open. BTC 5-minute markets cycling on screen. A green fill popped. She was behind me in line. Anthropic badge on her lanyard. "Wait. Is that Claude running trades" I told her. Claude Code. Four repos. $25 a month. She sat down across from me without asking. "I'm on the evals team. We benchmark Claude's reasoning on stuff way simpler than this. You're letting it run a full execution loop with real money" I showed her the first repo. 86 million trades. Claude finds which wallets consistently win. Reverse-engineers their timing. Copies their exits. "So you're not writing trading rules. Claude is finding them on its own from the data" Exactly. It watches Binance and Coinbase diverge from Chainlink. When both exchanges move $50+ the same way and Chainlink lags - it enters. 94% follow-through. Another trade closed on screen. +$41. "How often does it trade" About 55 times a day. Out of 400+ windows. Kills 85%. Only enters when order book imbalance confirms the price signal. Three commands. 500+ markets. Claude scores them in 20 minutes. She watched the screen for a while. My setup: Claude API - $20/mo VPS - $5/mo Everything else - free 30 days. 1,847 trades. 71% win rate. +$14,200. Copytrade here: She leaned back. "We spend all day testing what Claude can do in theory. You just showed me what it does in practice" I published the full breakdown that night. She DM'd me the next morning. "My team lead read your article. He wants to know if you'll consult for us" I told him to read it again. Everything's in there. Too late to gatekeep.

Lunar

68,586 просмотров • 2 месяцев назад