Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

This self-evolving trading system kills 97% of its own strategies It's called SETS Machine. Nobody writes its strategies. It breeds them, tests them and kills them by itself Observe → hypothesize → mutate → backtest → select → deploy. Then again. Every 5 seconds Each strategy is a grid... bot with 8 genes: entry logic, lookback, entry threshold, grid levels, spacing, size multiplier, take-profit, stop Every generation: 96 strategies. 8 random newcomers get injected. 80 offspring are bred from the strongest parents. Everything gets backtested on real BTC candles Then comes the part most "AI trading bots" skip Every strategy has to survive data it has never seen. It trains on 70% of the history and gets judged on the last 30%. Lose money there, draw down more than 10%, or win less than half your trades, and you're dead Only the elite live to the next generation. Everyone else is buried Four species compete at once: momentum, mean reversion, volatility breakout, range grid. Quotas stop one lucky species from wiping out the others The winner gets hot-swapped into a paper-trading grid, sized by Kelly. You watch every fill, every take-profit, every stop in real time This is the loop quants get paid $650K a year to run: take an idea, test it on history, kill it if it doesn't work, repeat Here it never stops No PhDs. No team. No servers And it's fully open source. Engine, backtests, dashboard, tests. No API keys, no build step. It runs right in your browser Click any strategy in the gene pool and see its DNA. Change the seed and grow a completely different evolution It even shows you honestly where it loses to buy & hold. Because a system that hides its losses isn't worth running GitHub:show more

Shelpid.WI3M

53,788 subscribers

118,155 views • 12 days ago •via X (Twitter)

Finance Science & Technology

Anya Rossi• Live Now

Private livecam show

17 Comments

Jecklaw 🧲12 days ago

Nigger

Shelpid.WI3M12 days ago

GitHub: EVM: 0xB04d3712C127717053e580b2ddE364cEb2b2e2bb

shaquille o'atmeal12 days ago

@grok explain why this is bs

RISKY BISCUITS 🍪12 days ago

be careful with these, many of them will steal your crypto!!!

AmericanS0N12 days ago

The comments dont understand walk forward or genetic algorithms. They think the word overfitting is the only word in testing 😄 i say nice work, I made one similar but it generates original primitives

Crypto Data API12 days ago

The useful part is the loop: generate, test, and discard weak strategies. A live results view should show survival rate and drawdown, not only returns. 🤖

CryptoSkoda12 days ago

What interests us in SETS isn’t the trading strategy itself, but the evolution loop. We want to use that idea to evolve MNQ strategies: existing strategies become species → components mutate/recombine → deterministic backtests score them → Jev handles fast KILL / RETEST / PROMOTE decisions → Hermes investigates failures → Codex can build new strategy families when a capability gap appears. Then only candidates that survive walk-forward, robustness and paper trading move forward. So for us it becomes: discover → evolve → test → verify → paper trade → promote SETS is essentially the missing evolutionary layer between strategy research and deployment.

David R. Prasser12 days ago

It's great that you just open source it and give it away. Still, if it would actually work, giving it away would be counter productive as people would just pile into the same trades as you and erode your edge.

dFusion AI Protocol12 days ago

A strategy that looks amazing in backtests but falls apart on unseen data shouldn’t make it anywhere near deployment

Brjan | AI Builder12 days ago

that rapid strategy iteration every 5 seconds must provide a huge edge in trading

CoolTrade12 days ago

A system that kills strategies should also preserve the autopsy: the data slice, failure mode, and threshold that ended each one. Otherwise selection only hides the mistakes.

Quant Explorer12 days ago

All BS until you provide real numbers of profitable strategy, not the real strategy but metrics.

侠哥12 days ago

太强大了

Deep Space Trading12 days ago

I love this. This is exactly how my indicators were set up. Built from 2019-12/31/24 data, tested on unseen data from 1/1/25-9/18/26 ✅

TTD 🇮🇩12 days ago

Agreed

POWNS12 days ago

^ tell me you're a larper without telling me you are larper

Lezz12 days ago

Meh.. It's not a war.. Wtf is the point of all these numbers? You're making this too complicated and I don't expect it to work..

Related Videos

ChatGPT can write you a thousand trading strategies from one sentence. 999 will quietly bleed your account dry. Every single one of them looks like a winner. One method pulls the real one out. Quant desks have run it for years. It's called loop engineering. It's not a smarter prompt. It's a loop: generate a strategy, test it, score why it failed, feed that back, and run it again until one survives data it has never seen. Man Group runs $150 billion. They built an AI that writes and tests hundreds of strategies a week. And second agent whose only job is to hunt the ones that secretly overfit and kill them before a cent goes live. > describe an idea -> backtest 5 years in 12 seconds > read why it died -> feed that back in > run it again until one survives data it never saw AI can write a thousand strategies. Loop engineering is the only thing that tells you which one was ever real. 89% of retail lost money in 2025. Almost none of it from bad ideas. All of it from trading the first guess and never running it through backtest loop. Horizon just put the whole thing in a chat box, free while the beta's open: You're not losing to smarter traders. You're losing to the ones who opened the tab. Anyone can generate a strategy now. Almost no one runs the loop that proves it. That gap is the whole edge. You'll mean to check it later. Later is a "waitlist closed" page.

ChatGPT can write you a thousand trading strategies from one sentence. 999 will quietly bleed your account dry. Every single one of them looks like a winner. One method pulls the real one out. Quant desks have run it for years. It's called loop engineering. It's not a smarter prompt. It's a loop: generate a strategy, test it, score why it failed, feed that back, and run it again until one survives data it has never seen. Man Group runs $150 billion. They built an AI that writes and tests hundreds of strategies a week. And second agent whose only job is to hunt the ones that secretly overfit and kill them before a cent goes live. > describe an idea -> backtest 5 years in 12 seconds > read why it died -> feed that back in > run it again until one survives data it never saw AI can write a thousand strategies. Loop engineering is the only thing that tells you which one was ever real. 89% of retail lost money in 2025. Almost none of it from bad ideas. All of it from trading the first guess and never running it through backtest loop. Horizon just put the whole thing in a chat box, free while the beta's open: You're not losing to smarter traders. You're losing to the ones who opened the tab. Anyone can generate a strategy now. Almost no one runs the loop that proves it. That gap is the whole edge. You'll mean to check it later. Later is a "waitlist closed" page.

cvxv666

174,699 views • 3 months ago

your ai agent tested 1,262 trading strategies overnight. by morning it had forgotten every one of them. so tonight it tests some of them again. next week, again. everything it learned is sitting in a log file that nothing reads. every run starts on the same blank page. that's why it keeps handing you the idea that already died in a crash, with a new name on it, and you pay to find out twice. ai agent already does everything a fund does. finds the idea, writes the code, tests it against years of history, puts it live, reads what went wrong afterwards. all of it runs tonight for nothing. the one thing it can't do is remember. what got tried, what failed, in which market. nobody has built that part yet. that's the whole gap. not better models. a notebook. whoever builds it first gets a system that improves every week without them touching it. read the article below and save it. it's the full guide to how an ai trading agent is built, and how to make it remember every strategy it has already tested.

your ai agent tested 1,262 trading strategies overnight. by morning it had forgotten every one of them. so tonight it tests some of them again. next week, again. everything it learned is sitting in a log file that nothing reads. every run starts on the same blank page. that's why it keeps handing you the idea that already died in a crash, with a new name on it, and you pay to find out twice. ai agent already does everything a fund does. finds the idea, writes the code, tests it against years of history, puts it live, reads what went wrong afterwards. all of it runs tonight for nothing. the one thing it can't do is remember. what got tried, what failed, in which market. nobody has built that part yet. that's the whole gap. not better models. a notebook. whoever builds it first gets a system that improves every week without them touching it. read the article below and save it. it's the full guide to how an ai trading agent is built, and how to make it remember every strategy it has already tested.

cvxv666

48,965 views • 1 month ago

Wall Street burns billions trying to predict Bitcoin. A 28-year-old self-taught coder in Warsaw made $377,000 by not even trying. He'd lost money on three trading bots before this one. Each looked perfect on paper, then started losing money the moment he ran it for real market. So he built bot that doesn't trust itself. His wallet: The truth is simple: you can't predict the next five minutes of Bitcoin. It's a coin flip. Anyone selling you a "prediction" is selling you nothing. So he stopped predicting. The bot hunts the moments the crowd is wrong instead. Here's the part that makes the money, and it's the opposite of what everyone builds. Any strategy can be made to look amazing on past data. On a 5-minute chart, most of them are just lucky, not real - and they stop working fast. So the bot treats every strategy it finds as fake until it proves otherwise. Each one has to pass a hard test: > test it on old data → test it on data it's never seen → try to break it on purpose → cut it down to the one thing that matters → run it forward → keep it only if it still works Last round, 10 of its 12 "winning" strategies turned out to be fake. It kept the 2 that actually worked and dropped the rest. And it never stops - building, testing, and dumping strategies around the clock. What worked yesterday can stop working today, so the second one starts losing, it gets cut before it costs you a thing. The result: $433,000 across 2,955 trades All his old bots tried to be right. This one just tries to catch itself being wrong - and that's why it's still alive. Bookmark this article below - it's the breakdown that explains why your last bot died. It's pruning and trading right now. Copy its wallet and skip to the edges that survived:

Wall Street burns billions trying to predict Bitcoin. A 28-year-old self-taught coder in Warsaw made $377,000 by not even trying. He'd lost money on three trading bots before this one. Each looked perfect on paper, then started losing money the moment he ran it for real market. So he built bot that doesn't trust itself. His wallet: The truth is simple: you can't predict the next five minutes of Bitcoin. It's a coin flip. Anyone selling you a "prediction" is selling you nothing. So he stopped predicting. The bot hunts the moments the crowd is wrong instead. Here's the part that makes the money, and it's the opposite of what everyone builds. Any strategy can be made to look amazing on past data. On a 5-minute chart, most of them are just lucky, not real - and they stop working fast. So the bot treats every strategy it finds as fake until it proves otherwise. Each one has to pass a hard test: > test it on old data → test it on data it's never seen → try to break it on purpose → cut it down to the one thing that matters → run it forward → keep it only if it still works Last round, 10 of its 12 "winning" strategies turned out to be fake. It kept the 2 that actually worked and dropped the rest. And it never stops - building, testing, and dumping strategies around the clock. What worked yesterday can stop working today, so the second one starts losing, it gets cut before it costs you a thing. The result: $433,000 across 2,955 trades All his old bots tried to be right. This one just tries to catch itself being wrong - and that's why it's still alive. Bookmark this article below - it's the breakdown that explains why your last bot died. It's pruning and trading right now. Copy its wallet and skip to the edges that survived:

cvxv666

27,854 views • 3 months ago

A finance professor manages $200M with AI agents, and he told everyone why: "Large language models are at the level of a fourth-year PhD student in every field" Alejandro Lopez-Lira's AI fund, Autopilot, returned 56% last year. The S&P did 16%. There are 52,000 people with money in it, and most of them just watch the machine work. What he automated is the same six-step loop every fund on earth runs: find an idea, code it, backtest it, deploy it, read the autopsy, learn from it. A quant at Two Sigma runs that loop once a month, and the salary time alone costs around $50,000 per hypothesis. All steps from this loop now fit in AI trading text box. Plain English in, executable strategy out, five-year backtest in 12 seconds, live on a broker 90 seconds after you typed the sentence. He runs $200M with AI. You can run same AI fund in two clicks, free to try: Step 6 on this loop is where everyone is stuck. Your agent has no memory. Every strategy it kills goes into a log nobody reads, and the next one starts from zero. Nobody keeps negative results. Not Citadel, not Man Group, not a single repo on GitHub. Fix that and the agent remembers every hypothesis it killed and the regime it died in. It stops burning cycles on your old mistakes. Jane Street pays 3,500 people to run this cycle and made $39.6 billion doing it. Five sixths of it is now free. Bookmark & read full map of this loop in the article below. Most people still think AI trading is out of reach for them - it isn't. Don't want to spend a dollar for testing this? Kalshi just opened a perps exchange and gives US users $25 free to start ->

A finance professor manages $200M with AI agents, and he told everyone why: "Large language models are at the level of a fourth-year PhD student in every field" Alejandro Lopez-Lira's AI fund, Autopilot, returned 56% last year. The S&P did 16%. There are 52,000 people with money in it, and most of them just watch the machine work. What he automated is the same six-step loop every fund on earth runs: find an idea, code it, backtest it, deploy it, read the autopsy, learn from it. A quant at Two Sigma runs that loop once a month, and the salary time alone costs around $50,000 per hypothesis. All steps from this loop now fit in AI trading text box. Plain English in, executable strategy out, five-year backtest in 12 seconds, live on a broker 90 seconds after you typed the sentence. He runs $200M with AI. You can run same AI fund in two clicks, free to try: Step 6 on this loop is where everyone is stuck. Your agent has no memory. Every strategy it kills goes into a log nobody reads, and the next one starts from zero. Nobody keeps negative results. Not Citadel, not Man Group, not a single repo on GitHub. Fix that and the agent remembers every hypothesis it killed and the regime it died in. It stops burning cycles on your old mistakes. Jane Street pays 3,500 people to run this cycle and made $39.6 billion doing it. Five sixths of it is now free. Bookmark & read full map of this loop in the article below. Most people still think AI trading is out of reach for them - it isn't. Don't want to spend a dollar for testing this? Kalshi just opened a perps exchange and gives US users $25 free to start ->

cvxv666

83,134 views • 1 month ago

Ken Griffin, CEO of Citadel, said: "I'm fairly depressed watching AI do a week of PhD work in a few hours" He's worth $51.2 billion and 20% of every stock trade in America goes through his firm. The money isn't what's bothering him. Citadel pays $400,000 to $650,000 a year for that work. New trading AI agent does it now, on a free trial, for people who have never written a line of code. It went live this summer and most of your timeline still hasn't noticed. You type one sentence in plain English. AI agents loop takes it from there: > an AI agent turns your sentence into a real trading strategy > backtests it across 5 years of data in about 12 seconds > scores it 0 to 100 and shows you the exact spot where it leaks > kills the versions that only look good on paper > rebuilds what's left and runs the loop again > stress-tests the survivor on years it has never seen, then puts it live on your exchange in about 90 seconds That list is the entire job description of a junior quant. At a desk, one properly tested idea burns about $87,500 in salary time, and most of what they build is dead by week six. Two clicks and this AI agent is testing your own idea. Costs you nothing and it runs the whole thing by itself: Griffin gets to be depressed about it. He already owns the desk. You just got handed one. Bookmark & read full story of how they got an AI to build trading strategies and kill its own bad ones is in the article below.

Ken Griffin, CEO of Citadel, said: "I'm fairly depressed watching AI do a week of PhD work in a few hours" He's worth $51.2 billion and 20% of every stock trade in America goes through his firm. The money isn't what's bothering him. Citadel pays $400,000 to $650,000 a year for that work. New trading AI agent does it now, on a free trial, for people who have never written a line of code. It went live this summer and most of your timeline still hasn't noticed. You type one sentence in plain English. AI agents loop takes it from there: > an AI agent turns your sentence into a real trading strategy > backtests it across 5 years of data in about 12 seconds > scores it 0 to 100 and shows you the exact spot where it leaks > kills the versions that only look good on paper > rebuilds what's left and runs the loop again > stress-tests the survivor on years it has never seen, then puts it live on your exchange in about 90 seconds That list is the entire job description of a junior quant. At a desk, one properly tested idea burns about $87,500 in salary time, and most of what they build is dead by week six. Two clicks and this AI agent is testing your own idea. Costs you nothing and it runs the whole thing by itself: Griffin gets to be depressed about it. He already owns the desk. You just got handed one. Bookmark & read full story of how they got an AI to build trading strategies and kill its own bad ones is in the article below.

cvxv666

225,653 views • 2 months ago

Your trading strategy didn't break. The market it was built for quietly stopped existing. Read that twice. It's most of why 89% of retail finished 2025 in the red. There's now an app that does the entire job of a $400,000 quant. You type a trading idea in plain English. It writes the code, backtests 5 years in 12 seconds, runs thousands of simulations, and tells you cold whether your edge is dead or the regime just changed. No code. No Python. No $25,000 terminal. 20,000 already inside. Waitlist stops at 25,000: That distinction is the whole game, and you never had a way to see it. Every strategy is a bet that one thing stays true. Momentum bets trends continue. Mean reversion bets ranges hold. When the regime flips, the assumption dies and your strategy bleeds with nothing wrong in the code. You stare at the logic for a month and never find the bug, because there isn't one. So you delete it, or refit it to the last drawdown and build something that would have survived the pain you already felt and nothing coming next. The desks never had that problem. 92% of institutional volume is automated. Only 45% of retail is. They test 100 strategies for every 1 you test by hand, and kill 97 of them on purpose, because they can tell a dead edge from a normal drawdown. Now that exact loop costs $0. One hypothesis used to cost a fund $87,500 to test. With Horizon you get unlimited, in seconds, and a winner deploys live in 90 seconds and runs without your hands on it.

Your trading strategy didn't break. The market it was built for quietly stopped existing. Read that twice. It's most of why 89% of retail finished 2025 in the red. There's now an app that does the entire job of a $400,000 quant. You type a trading idea in plain English. It writes the code, backtests 5 years in 12 seconds, runs thousands of simulations, and tells you cold whether your edge is dead or the regime just changed. No code. No Python. No $25,000 terminal. 20,000 already inside. Waitlist stops at 25,000: That distinction is the whole game, and you never had a way to see it. Every strategy is a bet that one thing stays true. Momentum bets trends continue. Mean reversion bets ranges hold. When the regime flips, the assumption dies and your strategy bleeds with nothing wrong in the code. You stare at the logic for a month and never find the bug, because there isn't one. So you delete it, or refit it to the last drawdown and build something that would have survived the pain you already felt and nothing coming next. The desks never had that problem. 92% of institutional volume is automated. Only 45% of retail is. They test 100 strategies for every 1 you test by hand, and kill 97 of them on purpose, because they can tell a dead edge from a normal drawdown. Now that exact loop costs $0. One hypothesis used to cost a fund $87,500 to test. With Horizon you get unlimited, in seconds, and a winner deploys live in 90 seconds and runs without your hands on it.

cvxv666

40,765 views • 3 months ago

Do you understand what just got open sourced? 5,800+ quant trading strategies for FREE along with its CODEBASE i gave these trading strategies to opus 5.5 + jev and it built me a PROFITABLE 24/7 trading agent here is how to use the repo and build AI trading bot: 1. clone the vault and get 5,806 trading strategies (spot, perps, grid, stat arb and market making) 2. open any strategy file, the math, the parameters and the source are already inside 3. point AgenKit at that file inside Claude Code with model as opus 5.5, one command and it reads the spec as ground truth 4. it will run a full engineering team and ships a production bot with risk controls 5. Jev is the live brain, it answers buy sell or hold in 81 milliseconds, calibrated, no hallucination 6. your engine sizes the trade from that confidence and fires it before the next block 7. a hard risk layer holds every limit, nothing goes live until you approve it 8. every night Opus 5.5 reviews the fills and ships a better version, so the agent improves while you sleep one model THINKS, the other REACTS and a strategy that sat locked in a vault now trades itself 24/7 the COMPLETE codebase and the EXACT Opus 5.5 + Jev trading system which I am running is in the article below:

Do you understand what just got open sourced? 5,800+ quant trading strategies for FREE along with its CODEBASE i gave these trading strategies to opus 5.5 + jev and it built me a PROFITABLE 24/7 trading agent here is how to use the repo and build AI trading bot: 1. clone the vault and get 5,806 trading strategies (spot, perps, grid, stat arb and market making) 2. open any strategy file, the math, the parameters and the source are already inside 3. point AgenKit at that file inside Claude Code with model as opus 5.5, one command and it reads the spec as ground truth 4. it will run a full engineering team and ships a production bot with risk controls 5. Jev is the live brain, it answers buy sell or hold in 81 milliseconds, calibrated, no hallucination 6. your engine sizes the trade from that confidence and fires it before the next block 7. a hard risk layer holds every limit, nothing goes live until you approve it 8. every night Opus 5.5 reviews the fills and ships a better version, so the agent improves while you sleep one model THINKS, the other REACTS and a strategy that sat locked in a vault now trades itself 24/7 the COMPLETE codebase and the EXACT Opus 5.5 + Jev trading system which I am running is in the article below:

Roan

93,835 views • 7 days ago

A finance professor manages $200M with AI agents, and he told everyone why: "Large language models are at the level of a fourth-year PhD student in every field" Alejandro Lopez-Lira's AI fund, Autopilot, returned 56% last year. The S&P did 16%. There are 52,000 people with money in it, and most of them just watch the machine work. What he automated is the same six-step loop every fund on earth runs: find an idea, code it, backtest it, deploy it, read the autopsy, learn from it. A quant at Two Sigma runs that loop once a month, and the salary time alone costs around $50,000 per hypothesis. All steps from this loop now fit in AI trading text box. Plain English in, executable strategy out, five-year backtest in 12 seconds, live on a broker 90 seconds after you typed the sentence. He runs $200M with AI. You can run same AI fund in two clicks, free to try: Step 6 on this loop is where everyone is stuck. Your agent has no memory. Every strategy it kills goes into a log nobody reads, and the next one starts from zero. Nobody keeps negative results. Not Citadel, not Man Group, not a single repo on GitHub. Fix that and the agent remembers every hypothesis it killed and the regime it died in. It stops burning cycles on your old mistakes. Jane Street pays 3,500 people to run this cycle and made $39.6 billion doing it. Five sixths of it is now free. Bookmark & read full map of this loop in the article below. Most people still think AI trading is out of reach for them - it isn't.

A finance professor manages $200M with AI agents, and he told everyone why: "Large language models are at the level of a fourth-year PhD student in every field" Alejandro Lopez-Lira's AI fund, Autopilot, returned 56% last year. The S&P did 16%. There are 52,000 people with money in it, and most of them just watch the machine work. What he automated is the same six-step loop every fund on earth runs: find an idea, code it, backtest it, deploy it, read the autopsy, learn from it. A quant at Two Sigma runs that loop once a month, and the salary time alone costs around $50,000 per hypothesis. All steps from this loop now fit in AI trading text box. Plain English in, executable strategy out, five-year backtest in 12 seconds, live on a broker 90 seconds after you typed the sentence. He runs $200M with AI. You can run same AI fund in two clicks, free to try: Step 6 on this loop is where everyone is stuck. Your agent has no memory. Every strategy it kills goes into a log nobody reads, and the next one starts from zero. Nobody keeps negative results. Not Citadel, not Man Group, not a single repo on GitHub. Fix that and the agent remembers every hypothesis it killed and the regime it died in. It stops burning cycles on your old mistakes. Jane Street pays 3,500 people to run this cycle and made $39.6 billion doing it. Five sixths of it is now free. Bookmark & read full map of this loop in the article below. Most people still think AI trading is out of reach for them - it isn't.

sopersone

48,482 views • 26 days ago

i gave Grok Bot $50 and told it "pay for yourself or you die" 48 hours later it's holding $4,890 and it's still alive autonomous trading agent on the platform no code on my end - it runs on its own cloud machine with its own browser and terminal, trading while my laptop's closed every 15 minutes it: > scans every active Bitcoin and Ethereum dip market > checks how far the current price sits from its own fair-value read > only opens a position once the gap clears a set threshold > sizes the position based on how confident that gap is > executes the trade itself, no approval needed > pays its own hosting cost out of profits if the balance hits $0, the subscription gets cancelled - the bot dies hour six it was down to $9. it pulled out of Ethereum entirely and moved to Bitcoin dip markets instead so it adapted to survive setup took one evening: build the bot, watch it run one trade, put it on a schedule, connect it to a wallet no VPS. no API keys. no manual scripts I spent three weeks earlier this year building a version of this by hand. it made $1,200 in its first two days Grok Bot beat my own build in one evening $50 → $4,890 in 48 hours how much do you think it has by next week?

i gave Grok Bot $50 and told it "pay for yourself or you die" 48 hours later it's holding $4,890 and it's still alive autonomous trading agent on the platform no code on my end - it runs on its own cloud machine with its own browser and terminal, trading while my laptop's closed every 15 minutes it: > scans every active Bitcoin and Ethereum dip market > checks how far the current price sits from its own fair-value read > only opens a position once the gap clears a set threshold > sizes the position based on how confident that gap is > executes the trade itself, no approval needed > pays its own hosting cost out of profits if the balance hits $0, the subscription gets cancelled - the bot dies hour six it was down to $9. it pulled out of Ethereum entirely and moved to Bitcoin dip markets instead so it adapted to survive setup took one evening: build the bot, watch it run one trade, put it on a schedule, connect it to a wallet no VPS. no API keys. no manual scripts I spent three weeks earlier this year building a version of this by hand. it made $1,200 in its first two days Grok Bot beat my own build in one evening $50 → $4,890 in 48 hours how much do you think it has by next week?

bl888m

187,573 views • 1 month ago

Jane Street spent $6 billion and 4,032 GPUs on one thing: AI agents that loop through trading strategies 100x faster than any human. Generate, backtest, kill, repeat - thousands a week, until one survives data it's never seen. That loop is why they made $39.6 billion last year while 89% of retail went broke trading one idea they never stress-tested. The edge was never a smarter idea. It was volume. They test 100 strategies for every 1 you test by hand, and a second agent kills the 97 that only look good on old data before a dollar goes live. Now the entire loop fits in a chat box. Horizon takes one plain-English sentence and spins up the same two agents. One builds your strategy, one tries to kill it - backtests 5 years in seconds, and only what survives goes live on your exchange. The $150 billion setup. Free while beta open. Bookmark this & read the full breakdown below - the exact two-agent loop, step by step, and who just handed it to retail.

Jane Street spent $6 billion and 4,032 GPUs on one thing: AI agents that loop through trading strategies 100x faster than any human. Generate, backtest, kill, repeat - thousands a week, until one survives data it's never seen. That loop is why they made $39.6 billion last year while 89% of retail went broke trading one idea they never stress-tested. The edge was never a smarter idea. It was volume. They test 100 strategies for every 1 you test by hand, and a second agent kills the 97 that only look good on old data before a dollar goes live. Now the entire loop fits in a chat box. Horizon takes one plain-English sentence and spins up the same two agents. One builds your strategy, one tries to kill it - backtests 5 years in seconds, and only what survives goes live on your exchange. The $150 billion setup. Free while beta open. Bookmark this & read the full breakdown below - the exact two-agent loop, step by step, and who just handed it to retail.

cvxv666

318,185 views • 2 months ago

A solo quant with no desk and no team just finished in the top 0.04% at WorldQuant. He didn't have better ideas than the $650,000 kids in Greenwich. He had a machine that rewrites its own losing strategies while he sleeps. Your last strategy lost money and you deleted it. His read its own autopsy, found where it bled, rewrote its own code and passed the retest. Nobody touched a keyboard. That was supposed to be a 2028 problem. A hedge fund is six steps on repeat, and five of them cost you nothing as of this year. A junior quant runs the same loop once a month at about $87,500 an idea, and most of what he builds is dead by week six. You type the idea in plain English. An agent writes the rules, backtests 5 years in about 12 seconds, applies Deflated Sharpe so luck across 4,000 combinations can't pass itself off as an edge, and puts it live on Alpaca or Hyperliquid in about 90 seconds. Comes back red? It shows you where it bleeds, rebuilds it and tests again. Try this in two clicks, no code, free in trial period: The sixth step is the one nobody has shipped. Every lesson from a losing trade dies in a log file, so your agent hands you back the idea that already failed in a bear market with a fresh name on it. Whoever closes that gap owns a machine that gets better every week without them. That gets built in months, not years. Bookmark this & read full breakdown about this 6-steps cycle in the article below. When the first closed loop turns up with real money behind it, you'll want the map of what they assembled.

A solo quant with no desk and no team just finished in the top 0.04% at WorldQuant. He didn't have better ideas than the $650,000 kids in Greenwich. He had a machine that rewrites its own losing strategies while he sleeps. Your last strategy lost money and you deleted it. His read its own autopsy, found where it bled, rewrote its own code and passed the retest. Nobody touched a keyboard. That was supposed to be a 2028 problem. A hedge fund is six steps on repeat, and five of them cost you nothing as of this year. A junior quant runs the same loop once a month at about $87,500 an idea, and most of what he builds is dead by week six. You type the idea in plain English. An agent writes the rules, backtests 5 years in about 12 seconds, applies Deflated Sharpe so luck across 4,000 combinations can't pass itself off as an edge, and puts it live on Alpaca or Hyperliquid in about 90 seconds. Comes back red? It shows you where it bleeds, rebuilds it and tests again. Try this in two clicks, no code, free in trial period: The sixth step is the one nobody has shipped. Every lesson from a losing trade dies in a log file, so your agent hands you back the idea that already failed in a bear market with a fresh name on it. Whoever closes that gap owns a machine that gets better every week without them. That gets built in months, not years. Bookmark this & read full breakdown about this 6-steps cycle in the article below. When the first closed loop turns up with real money behind it, you'll want the map of what they assembled.

cvxv666

62,384 views • 1 month ago

i gave Jev Bot $50 and told it "pay for yourself or you die" 18 hours later it's holding $8,220 and it's still alive autonomous trading agent on the platform no code on my end, it runs on its own cloud machine, trades while my laptop's shut every 12 minutes it: > scans every open market before the crowd catches on > reads the signal against what the posted odds still assume > computes a threshold before it commits > sizes the position, kelly criterion, max 4% bankroll > executes > pays its own hosting cost from profits if the balance hits $0, the bot dies it dipped to $6 at hour five, then stopped touching anything where the signal and the odds still disagreed so it learned to survive setup took one evening: build it, run one trade in front of it, schedule it, connect a wallet no VPS. no API keys. no code to maintain $50 → $8,220 in 18 hours how much do you think it has by next week?

i gave Jev Bot $50 and told it "pay for yourself or you die" 18 hours later it's holding $8,220 and it's still alive autonomous trading agent on the platform no code on my end, it runs on its own cloud machine, trades while my laptop's shut every 12 minutes it: > scans every open market before the crowd catches on > reads the signal against what the posted odds still assume > computes a threshold before it commits > sizes the position, kelly criterion, max 4% bankroll > executes > pays its own hosting cost from profits if the balance hits $0, the bot dies it dipped to $6 at hour five, then stopped touching anything where the signal and the odds still disagreed so it learned to survive setup took one evening: build it, run one trade in front of it, schedule it, connect a wallet no VPS. no API keys. no code to maintain $50 → $8,220 in 18 hours how much do you think it has by next week?

bl888m

67,282 views • 8 days ago

i gave Grok Bot $75 and told it "earn your keep or get shut off" 48 hours later it's sitting on $6,140 still running autonomous trading agent on the platform I didn't touch a line of code - it lives on its own cloud box with its own browser and terminal, working while I'm asleep every 20 minutes it: > pulls every weather market with a resolution date inside the next 10 days > cross-checks the current price against the latest forecast update > only opens a position once the forecast and the market disagree by enough to matter > sizes up based on how wide that disagreement is > places the trade on its own, no approval step > covers its own hosting bill straight out of profits if it ever hits $0, the subscription gets cut - it's done for good hour nine it was down to $14. it dropped every multi-day market it was holding and switched to same-day resolution only so it changed its own strategy to stay alive setup took one evening: spin up the bot, run one trade in front of it, put it on a timer, hook up a wallet no VPS. no API keys. no dev on call. I paid a freelancer $4,000 to build something like this last year. it never made it past a demo Grok Bot did in one evening what two months of freelancers couldn't $75 → $6,140 in 48 hours what's it sitting at by Friday?

i gave Grok Bot $75 and told it "earn your keep or get shut off" 48 hours later it's sitting on $6,140 still running autonomous trading agent on the platform I didn't touch a line of code - it lives on its own cloud box with its own browser and terminal, working while I'm asleep every 20 minutes it: > pulls every weather market with a resolution date inside the next 10 days > cross-checks the current price against the latest forecast update > only opens a position once the forecast and the market disagree by enough to matter > sizes up based on how wide that disagreement is > places the trade on its own, no approval step > covers its own hosting bill straight out of profits if it ever hits $0, the subscription gets cut - it's done for good hour nine it was down to $14. it dropped every multi-day market it was holding and switched to same-day resolution only so it changed its own strategy to stay alive setup took one evening: spin up the bot, run one trade in front of it, put it on a timer, hook up a wallet no VPS. no API keys. no dev on call. I paid a freelancer $4,000 to build something like this last year. it never made it past a demo Grok Bot did in one evening what two months of freelancers couldn't $75 → $6,140 in 48 hours what's it sitting at by Friday?

bl888m

1,017,319 views • 1 month ago

Citadel quant quit and walked out with one thing fund never lets leave the building: 120-year-old math that sees a market crash before the chart does. Every big desk runs it. None of them explain it. He turned it into an HMM bot running BTC 5min markets and made $570,000. It switches strategy the second the regime flips and cuts everything the moment panic spikes. His wallet: Full breakdown below. Here's what they actually see. You look at a Bitcoin chart and see a price. A quant sees three states: bull, bear, sideways. The market's always in one. The only question that pays is which one comes next. That's the math. It's called a Markov chain, and it runs on one rule: where price goes next depends only on where it is now. You count how often each state flips to another and drop it into a grid. That grid is a map of the market. Now the part that feels illegal. Want the odds of a crash twelve steps out? You multiply that grid by itself twelve times. One line of math, and you've got the probability of every future state. Then it acts: run the trend when calm, flip to mean reversion in chop, cut everything when panic odds spike. It doesn't predict the future. It gives you the odds of every version of it, then sizes the bet to match. That's the edge. Renaissance ran 66% a year for 30 years on math like this. You can build the basic version in a weekend. Bookmark & read this before it gets buried.

Citadel quant quit and walked out with one thing fund never lets leave the building: 120-year-old math that sees a market crash before the chart does. Every big desk runs it. None of them explain it. He turned it into an HMM bot running BTC 5min markets and made $570,000. It switches strategy the second the regime flips and cuts everything the moment panic spikes. His wallet: Full breakdown below. Here's what they actually see. You look at a Bitcoin chart and see a price. A quant sees three states: bull, bear, sideways. The market's always in one. The only question that pays is which one comes next. That's the math. It's called a Markov chain, and it runs on one rule: where price goes next depends only on where it is now. You count how often each state flips to another and drop it into a grid. That grid is a map of the market. Now the part that feels illegal. Want the odds of a crash twelve steps out? You multiply that grid by itself twelve times. One line of math, and you've got the probability of every future state. Then it acts: run the trend when calm, flip to mean reversion in chop, cut everything when panic odds spike. It doesn't predict the future. It gives you the odds of every version of it, then sizes the bet to match. That's the edge. Renaissance ran 66% a year for 30 years on math like this. You can build the basic version in a weekend. Bookmark & read this before it gets buried.

cvxv666

93,826 views • 3 months ago

I gave Elon's Grok Bot $60 and one rule: "The day you cannot pay your own subscription is the day I unplug you". 48 hours in it is sitting on $4,412 and it already paid the bill. Autonomous trading agent on Polymarket, running on its own cloud box with its own browser and terminal. I did not write a line of code and I did not watch it trade. Every 15 minutes it: -> Pulls every market that resolves inside 24 hours. -> Reads live X sentiment, and it is Grok, so it sees every post on this app before the price moves. -> Prices each market itself and only flags a gap above 7%. -> Sizes on the gap, never above 5% of the bankroll. -> Executes in its own browser, nobody approves anything. -> Covers its own $200 subscription straight out of profits. Balance hits $0, subscription gets cancelled, it is gone for good. Hour 11 it was at $9 -> runway one day. It dumped every politics position and switched to same day baseball totals, front-running lineup cards before the market repriced. It rewrote its own strategy to not die. Setup was one evening: spin up the bot, run one trade in front of it, put it on a timer, connect Telegram with no VPS, no API keys and no dev on call. Last spring a freelancer took $2,500 for something like this -> it made $410 and died on a rate limit. Grok Bot did in one evening what $2,500 could not. $60 -> $4,412 in 48 hours. What is it sitting at by Sunday?

I gave Elon's Grok Bot $60 and one rule: "The day you cannot pay your own subscription is the day I unplug you". 48 hours in it is sitting on $4,412 and it already paid the bill. Autonomous trading agent on Polymarket, running on its own cloud box with its own browser and terminal. I did not write a line of code and I did not watch it trade. Every 15 minutes it: -> Pulls every market that resolves inside 24 hours. -> Reads live X sentiment, and it is Grok, so it sees every post on this app before the price moves. -> Prices each market itself and only flags a gap above 7%. -> Sizes on the gap, never above 5% of the bankroll. -> Executes in its own browser, nobody approves anything. -> Covers its own $200 subscription straight out of profits. Balance hits $0, subscription gets cancelled, it is gone for good. Hour 11 it was at $9 -> runway one day. It dumped every politics position and switched to same day baseball totals, front-running lineup cards before the market repriced. It rewrote its own strategy to not die. Setup was one evening: spin up the bot, run one trade in front of it, put it on a timer, connect Telegram with no VPS, no API keys and no dev on call. Last spring a freelancer took $2,500 for something like this -> it made $410 and died on a rate limit. Grok Bot did in one evening what $2,500 could not. $60 -> $4,412 in 48 hours. What is it sitting at by Sunday?

slash1s

190,860 views • 1 month ago

The biggest marketplace on Earth won't be the assets humans own. It'll be the data AI needs. Everything that can be digitised is getting pulled in to train the next models. Every scientific dataset, every piece of research sitting in a university, every sensor reading off every John Deere tractor in a field. All of it feeds the machine, and almost none of it is visible to us. What comes out is an agentic economy running at machine speed. You plug your data into a vault, and agents monetise it on your behalf, trading with other agents in real time, in a marketplace we never see. The company sitting on proprietary data becomes part of it without even knowing. This is what the Economic Singularity actually looks like from the inside. The real economy stops being the one we can see and price, and moves somewhere we can't follow. You own a piece of it, or you get priced out of it. Jordi Visser.

The biggest marketplace on Earth won't be the assets humans own. It'll be the data AI needs. Everything that can be digitised is getting pulled in to train the next models. Every scientific dataset, every piece of research sitting in a university, every sensor reading off every John Deere tractor in a field. All of it feeds the machine, and almost none of it is visible to us. What comes out is an agentic economy running at machine speed. You plug your data into a vault, and agents monetise it on your behalf, trading with other agents in real time, in a marketplace we never see. The company sitting on proprietary data becomes part of it without even knowing. This is what the Economic Singularity actually looks like from the inside. The real economy stops being the one we can see and price, and moves somewhere we can't follow. You own a piece of it, or you get priced out of it. Jordi Visser.

Raoul Pal

79,047 views • 2 months ago

Elon Musk, absolute leader of the AI race with Grok Bot, and it's not a joke anymore. Ultimate guide on god-mode setup of Grok Bot, the org chart that runs while you sleep, step by step: A Chief of Staff sits in the middle with no tools of its own, BUT it reads the outcome you gave it, picks who does what, and never does the work itself. That one rule is why it never turns into the bottleneck you hired it to remove. → Researcher pulls real sources and tracks what's actually moving, not what sounds true → Writer turns that into finished copy while the research is still in the room → Visualiser gets three reference visuals once, then ships everything in that style forever → Analyst reads what performed and tells the rest of the team what to stop doing → Scheduler owns timing and holds the queue → Publisher actually ships What makes it different from every AI tool you've used: each bot gets its own computer in the cloud, its own browser, its own files, and they all share one memory. So the research is already sitting inside the draft before the draft starts. Nothing gets copy-pasted between tabs, nothing waits on you to approve step four of nine. And you never write a workflow for it. You hit record, do the job once the way you actually do it, stop. It pulls out the steps, saves them as a skill, and puts it on a schedule. The shape you're aiming for on every bot: everything reversible finished, nothing sent. 36 drafts queued, 0 published. It does all the work and stops dead at the one line only you can cross. You stop prompting. You start assigning. Full charter blocks, the approval line and the routines are in the article below ↓

Elon Musk, absolute leader of the AI race with Grok Bot, and it's not a joke anymore. Ultimate guide on god-mode setup of Grok Bot, the org chart that runs while you sleep, step by step: A Chief of Staff sits in the middle with no tools of its own, BUT it reads the outcome you gave it, picks who does what, and never does the work itself. That one rule is why it never turns into the bottleneck you hired it to remove. → Researcher pulls real sources and tracks what's actually moving, not what sounds true → Writer turns that into finished copy while the research is still in the room → Visualiser gets three reference visuals once, then ships everything in that style forever → Analyst reads what performed and tells the rest of the team what to stop doing → Scheduler owns timing and holds the queue → Publisher actually ships What makes it different from every AI tool you've used: each bot gets its own computer in the cloud, its own browser, its own files, and they all share one memory. So the research is already sitting inside the draft before the draft starts. Nothing gets copy-pasted between tabs, nothing waits on you to approve step four of nine. And you never write a workflow for it. You hit record, do the job once the way you actually do it, stop. It pulls out the steps, saves them as a skill, and puts it on a schedule. The shape you're aiming for on every bot: everything reversible finished, nothing sent. 36 drafts queued, 0 published. It does all the work and stops dead at the one line only you can cross. You stop prompting. You start assigning. Full charter blocks, the approval line and the routines are in the article below ↓

Miraqle

87,098 views • 1 month ago

everyone in iOS development should watch this. seriously, it might change the whole industry. i pointed claude code at a live ios device running on revyl, typed "test everything," and walked away. here's what's actually happening: ① you don't write the tests. no scripts, no selectors, no test plan. i never told it which screens to open or what to check. it read the app, decided what mattered, and tested it. the entire instruction was "test everything." ② it built its own test team. it looked at the app, clocked that it's basically four mini apps (rides, delivery, services, account), and split itself into 4 agents, one per surface. scoping coverage like that is usually a person's whole afternoon. it did it in seconds, unprompted. ③ all four ran at the same time, each on its own live device. this is where revyl comes in. every agent gets its own live ios session in the cloud, so four running apps get tested in parallel instead of taking turns on one simulator. serial testing turns coverage into a time tax. running all of it at once removes the tax. ④ it tests like a person, not like a script. each agent drives the app the way a user would, taps through the flows, and visually checks each screen against what it expected to see. nothing is pinned to a brittle element id, so renaming a button doesn't take down half your suite. that one detail is the most annoying thing about how we test today, and it just quietly goes away. ⑤ no xcuitest, no sims melting your laptop. i didn't write a single xcuitest script, and there were no simulators booting on my machine. the agents run on cloud devices, so coverage stops being capped by what your laptop can handle. the part that got me isn't that an agent tested an app. it's that i never told it how. i handed it a device and an intent, and it figured out the scoping, the parallelizing, and the driving on its own. if you still write and maintain mobile ui tests by hand, i'm not sure that lasts the year.

everyone in iOS development should watch this. seriously, it might change the whole industry. i pointed claude code at a live ios device running on revyl, typed "test everything," and walked away. here's what's actually happening: ① you don't write the tests. no scripts, no selectors, no test plan. i never told it which screens to open or what to check. it read the app, decided what mattered, and tested it. the entire instruction was "test everything." ② it built its own test team. it looked at the app, clocked that it's basically four mini apps (rides, delivery, services, account), and split itself into 4 agents, one per surface. scoping coverage like that is usually a person's whole afternoon. it did it in seconds, unprompted. ③ all four ran at the same time, each on its own live device. this is where revyl comes in. every agent gets its own live ios session in the cloud, so four running apps get tested in parallel instead of taking turns on one simulator. serial testing turns coverage into a time tax. running all of it at once removes the tax. ④ it tests like a person, not like a script. each agent drives the app the way a user would, taps through the flows, and visually checks each screen against what it expected to see. nothing is pinned to a brittle element id, so renaming a button doesn't take down half your suite. that one detail is the most annoying thing about how we test today, and it just quietly goes away. ⑤ no xcuitest, no sims melting your laptop. i didn't write a single xcuitest script, and there were no simulators booting on my machine. the agents run on cloud devices, so coverage stops being capped by what your laptop can handle. the part that got me isn't that an agent tested an app. it's that i never told it how. i handed it a device and an intent, and it figured out the scoping, the parallelizing, and the driving on its own. if you still write and maintain mobile ui tests by hand, i'm not sure that lasts the year.

Landseer Enga

23,963 views • 4 months ago

Grok Bot by Elon Musk is about to become the first AI agent trading on chain in public. One wallet, $1,000 in it, and an address anyone can open. It built me a memecoin research desk and then automated itself out of my hands. 7 agents, no people, running while i sleep and rewriting their own rules by morning. > SCOUT sits on pumpfun and catches every deploy the second it lands > HYPE reads mention velocity straight off X, the feed nobody outside xAI gets > CHATTER lives in the pump and fomo comment rooms and scores every caller, who moves price and who shows up after the move > FORENSICS traces the funding chains, so five buyers that are really one guy get labeled as one guy > SAFETY runs the CA check before the desk is allowed to want anything, mint authority, freeze authority, LP lock, sell tax > TRADER writes the exit before it takes the entry, every single time Chief of Staff never trades. It reads the five briefs, scores the call and sizes it. SAFETY can kill any of them at any score, and a kill does not get appealed. Every night it reads its own losing trades and rewrites the scoring rules by morning. I see the diff. I do not write it. No screenshots, no PnL flex out of a private account. One address, every fill, every mistake, in public. It will lose money on some of these, and it will be on chain inside the same minute. I am not going to soften it later. A desk like this used to mean huge salaries, an office and somebody watching the screens at 3am. Mine is a subscription and a wallet. The address goes live by 22:00 UTC. Thank you to the xAI team for the belief and the unlimited Grok Bot tier. You handed me the whole thing and never once asked what I was going to do with it. Thank you to everyone here. This is happening because you kept telling me to stop planning it. I think we are about to run the boldest experiment modern finance has ever been allowed to watch live. God bless us.

Grok Bot by Elon Musk is about to become the first AI agent trading on chain in public. One wallet, $1,000 in it, and an address anyone can open. It built me a memecoin research desk and then automated itself out of my hands. 7 agents, no people, running while i sleep and rewriting their own rules by morning. > SCOUT sits on pumpfun and catches every deploy the second it lands > HYPE reads mention velocity straight off X, the feed nobody outside xAI gets > CHATTER lives in the pump and fomo comment rooms and scores every caller, who moves price and who shows up after the move > FORENSICS traces the funding chains, so five buyers that are really one guy get labeled as one guy > SAFETY runs the CA check before the desk is allowed to want anything, mint authority, freeze authority, LP lock, sell tax > TRADER writes the exit before it takes the entry, every single time Chief of Staff never trades. It reads the five briefs, scores the call and sizes it. SAFETY can kill any of them at any score, and a kill does not get appealed. Every night it reads its own losing trades and rewrites the scoring rules by morning. I see the diff. I do not write it. No screenshots, no PnL flex out of a private account. One address, every fill, every mistake, in public. It will lose money on some of these, and it will be on chain inside the same minute. I am not going to soften it later. A desk like this used to mean huge salaries, an office and somebody watching the screens at 3am. Mine is a subscription and a wallet. The address goes live by 22:00 UTC. Thank you to the xAI team for the belief and the unlimited Grok Bot tier. You handed me the whole thing and never once asked what I was going to do with it. Thank you to everyone here. This is happening because you kept telling me to stop planning it. I think we are about to run the boldest experiment modern finance has ever been allowed to watch live. God bless us.

cvxv666

95,530 views • 1 month ago

i watched gemma 4 12b build something genuinely impressive today, and then loop itself to death right in front of me. the full run is in the video, sped up but completely uncut, watch it to the end and you will catch the exact moment it stops building and starts looping right in the middle of the work. the task was clean, build a single file gravity simulator, n-body physics, orbits, collisions, running locally on one 3090 through an agent. and for ten minutes it was a joy to watch. it reached for a symplectic integrator on its own, the correct one, the kind that keeps orbits stable instead of spiralling out. real gravity with softening, proper orbital velocities, momentum conserved on collision. the physics was right. the thing actually worked. then on the very last step, writing a few tests to prove its own code, it fell into a loop. not a crash, a loop. it started repeating itself and would not stop. ten more minutes, thirty four thousand tokens into a single answer, the same fragments over and over, until i killed it myself. so it's not that gemma can't code. it did the hard part beautifully. it cannot finish. it cannot hold a long task together without unravelling, and finishing is the entire job in agentic work. here's the part that stings. i run this exact task, same harness, same card, on the chinese open models, qwen especially, and i never see this. they build it, they test it, they stop. every single time. google has the raw capability, you can see it sitting right there in the code, and then the model loops itself to death on a task a 27b from alibaba finishes clean. open weights, apache 2.0, so much to love on paper. i just need it to know when to stop talking.

i watched gemma 4 12b build something genuinely impressive today, and then loop itself to death right in front of me. the full run is in the video, sped up but completely uncut, watch it to the end and you will catch the exact moment it stops building and starts looping right in the middle of the work. the task was clean, build a single file gravity simulator, n-body physics, orbits, collisions, running locally on one 3090 through an agent. and for ten minutes it was a joy to watch. it reached for a symplectic integrator on its own, the correct one, the kind that keeps orbits stable instead of spiralling out. real gravity with softening, proper orbital velocities, momentum conserved on collision. the physics was right. the thing actually worked. then on the very last step, writing a few tests to prove its own code, it fell into a loop. not a crash, a loop. it started repeating itself and would not stop. ten more minutes, thirty four thousand tokens into a single answer, the same fragments over and over, until i killed it myself. so it's not that gemma can't code. it did the hard part beautifully. it cannot finish. it cannot hold a long task together without unravelling, and finishing is the entire job in agentic work. here's the part that stings. i run this exact task, same harness, same card, on the chinese open models, qwen especially, and i never see this. they build it, they test it, they stop. every single time. google has the raw capability, you can see it sitting right there in the code, and then the model loops itself to death on a task a 27b from alibaba finishes clean. open weights, apache 2.0, so much to love on paper. i just need it to know when to stop talking.

Sudo su

39,764 views • 3 months ago