Fromsay (@FromsayAI) header
Fromsay (@FromsayAI) profile picture

Al, robotics | Demos, experiments, terminals & workflows | Prompts, tools & breakdowns

Joined January 2025
Photo by Fromsay (@FromsayAI)Photo by Fromsay (@FromsayAI)Photo by Fromsay (@FromsayAI)Photo by Fromsay (@FromsayAI)Photo by Fromsay (@FromsayAI)Photo by Fromsay (@FromsayAI)Photo by Fromsay (@FromsayAI)Photo by Fromsay (@FromsayAI)Photo by Fromsay (@FromsayAI)Photo by Fromsay (@FromsayAI)
Fromsay retweeted
Corinna Kopf says a lot of OnlyFans creators are clickbaiting their own fans: she paid $150 for a locked post and got a bra and panties She wasn't a fan. She was doing research. Weeks after launching her own page, which she says brought in a million dollars in a single day, the streamer bought other big creators' pay-per-view posts to see what they were really selling Here is how the trick works, in her words: "You can sell pay-per-view content where you make a caption, and then you see the picture." So she bought the pictures. "I was mind-boggled," she says, "because the caption would be like, full frontal. And then you'd unlock the picture and it's the laziest bra you've ever seen." The $150 post, she says, was the same: "they were literally in panties and a really super lacy bra. You got clickbaited." She is careful to say it isn't about what people post. "You should be able to post whatever the f*** you want," she says, "but don't say you're posting this and then not post it." And then the line that matters: "There's a lot of girls doing that." Look at the economics. The buyer pays before they can inspect the product. You can try on a shirt, read a sample chapter, watch a trailer. Here the only information is the caption, and the caption is written by the person who gets paid if you believe it. At that point the label stops being information and becomes advertising Run the arithmetic. If a creator sells one locked picture at $150 and 1,000 subscribers unlock it, that is $150,000 from a single post, and also 1,000 people who just learned what that creator's captions are worth. Corinna says she expected about 20,000 subscribers at launch and got "60-something thousand," so the pool is big enough to lose the first thousand and keep growing. That is why the trick survives: the cost of the lie is spread across a market too large to notice it My read is that it's the same thing that happened to YouTube thumbnails and news headlines. Any platform that rewards the click fills up with promises the content can't keep, and the first people to suffer are the ones who paid for the promise. Corinna, who says she posts a lot more than most people post, isn't making a moral argument. She's making a business one: the day your subscribers learn your caption and your picture don't match, you stop having a product In any market where the label is the product, the most valuable thing you can own is that people believe it Is that smart marketing, or just a scam with better lighting?
3
2
14
967
Fromsay retweeted
MY EXECUTOR AGENT TRIED TO GO ALL IN TWICE, MY RISK AGENT SAID NO BOTH TIMES, AND ONE OF THOSE COINS WENT +320% Four Grok Bot agents ran a $100 desk from 9 PM with one rule: never more than 28% on one coin. About 12:18 AM the Analyst scored PUPS at 96 and the Executor pulled the size to 100%. Risk denied it. The coin ran +320%, and the rule cost $481.89 on that single trade. The 4 agents: Scanner (blue) → finds new coins, skips the junk Analyst (pink) → scores the setup, marks the resistance Risk (amber) → owns the size, blocks the rest Executor (violet) → clicks Buy and Sell, argues a lot Night report: 11 trades, 7 wins, 4 stops, 20 coins skipped, $100 → $1,904.60. At 2:48 AM the Executor came back with CAW, a score of 91 and the same ask: all in. Same rule, same denial. CAW rugged −88%, and the 28% cap turned a −$534.65 disaster into −$149.69. Watch the toast right after the sell. Same rule, same agent, same argument. It cost me on the first coin and saved me on the second, and I couldn’t have known which was which. Which agent would you listen to, Executor who wants all in or Risk who says no?
1
1
15
350
MY EXECUTOR AGENT TRIED TO GO ALL IN TWICE, MY RISK AGENT SAID NO BOTH TIMES, AND ONE OF THOSE COINS WENT +320% Four Grok Bot agents ran a $100 desk from 9 PM with one rule: never more than 28% on one coin. About 12:18 AM the Analyst scored PUPS at 96 and the Executor pulled the size to 100%. Risk denied it. The coin ran +320%, and the rule cost $481.89 on that single trade. The 4 agents: Scanner (blue) → finds new coins, skips the junk Analyst (pink) → scores the setup, marks the resistance Risk (amber) → owns the size, blocks the rest Executor (violet) → clicks Buy and Sell, argues a lot Night report: 11 trades, 7 wins, 4 stops, 20 coins skipped, $100 → $1,904.60. At 2:48 AM the Executor came back with CAW, a score of 91 and the same ask: all in. Same rule, same denial. CAW rugged −88%, and the 28% cap turned a −$534.65 disaster into −$149.69. Watch the toast right after the sell. Same rule, same agent, same argument. It cost me on the first coin and saved me on the second, and I couldn’t have known which was which. Which agent would you listen to, Executor who wants all in or Risk who says no?
1
1
15
350
Fromsay retweeted
MY BOT LOCKED HALF ITS PROFITS IN A VAULT OVERNIGHT. IT COST ME $476. I'D DO IT AGAIN. $100 at 9 PM. Five Grok Bot agents, five cursors, one new rule: every time the trading desk doubles, RISK moves half of the profit into a vault. The agents can't touch it. Not for a better coin, not for a "sure thing", not for anything. 10:52 PM. LUNARA closes +$69.50. The desk hits $208. RISK moves $54.25 to the vault. 2:02 AM. WADDL closes. The desk doubles again. $107.25 goes in. 2:31 AM. BEARO scores 58, top 10 wallets hold 61%. FORENSIC says no. 13 minutes later it's down 52%. 4:31 AM. ROCKIT pays +$448.72. $329.78 goes to the vault. Then the last hour went bad. Three stops in a row: TOADLY, PIGGO, LUNARA. Here's the part I didn't expect. Without the vault, the same night ends at $1,458.17. Bigger bets after every double, bigger wins. The vault cost me $476. But those three stops cost $100.18 with the vault and would have cost $297.43 without it. And on a night where the last hour is five stops instead of three, the vault is the only reason there's anything left. Final: $982.38. $491.10 on the desk, $491.28 in the vault that nobody can trade. Most bots are built to make the number go up. This rule is built so the number can't all go back down. Would you lock half your profit, or let it all ride?
1
3
24
578
Out of 2,000 inmates, three men were taken to the hole after a 1968 prison riot, all three facing gas chamber offenses, and Danny Trejo was one of them. He did not expect to get out alive. A little over a year later, he walked out of prison a free man. In his Modern Wisdom interview, Trejo walks through the worst stretch of his life, and it starts with a joke only he could tell. "Cinco de Mayo to a real Mexican, spelled backwards, is get bail money." It was Cinco de Mayo, 1968, tequila inside the walls, and a riot broke out. "When a riot starts, it is like an explosion." When it was over, the prison wanted names. Trejo says it was alleged he threw a rock and hit Lieutenant Givens. "There's no proof. Somebody said that." Another man, Henry, kicked coach Stromat. "But all three gas chamber offenses." Out of 2,000 inmates, they took three of them to the hole. He remembers the cell. Naked, scared, asking God what happened to all that potential. He says he remembers not wanting to die, "screaming and yelling and begging and peeing my pants." Then he stopped bargaining for his life and asked for something smaller: "Dear heavenly Father, just let me die with dignity. I'll say your name every day. And I will do whatever I can for my fellow inmate." By his own account, the case did not end where he expected. "D.A. rejected it. Okay, get this out of here." A month before his release, the parole agent on the board, Mad Dog Madden, told him: "Trey, we're going to give you a chance to get out of here and bring us back a life sentence." Trejo only had a 10-year top, and he says he'll never forget that. Thirty days later, on August 23, 1969, he walked out of prison. "I was shocked." And then the part that stayed with me. He says he never made a deal with God for a good job or for an easy life. The only promise he made in that cell was to help his fellow inmates, and he kept it. A couple of months ago he asked God how he was doing. The answer: "You're almost out of hell, Trey." He treats everything since then as borrowed time. "It's like, take me now. My kids are great. I got great insurance. They'll be all right." And the line he ends on: "Everything good that has ever happened to me has happened as a direct result of helping someone else." Look at what he actually asked for. Not freedom, not money, not a second chance. Just dignity, and a chance to be useful to the men around him. He walked out with a 10-year top instead of a death sentence, and he spent the next decades doing exactly what he promised on a cell floor. Do you keep a promise made in the hole, or forget it the moment the door opens?
4
2
16
690
Fromsay retweeted
I GAVE 5 AI AGENTS $100 AND ONE RULE: 2 LOSSES IN A ROW AND THE DESK LOCKS FOR 30 MINUTES. BY 6am THEY HAD $1,330.16. Five Grok Bot agents ran the desk from 9pm, each with its own cursor. Every coin needs 5 votes in a row: SCOUT finds it, FORENSIC scores it, SIZER sizes it, RISK sets the stop, CHIEF clicks Buy. One no and nothing happens. They could pick any coin, any entry, any exit. The one thing they couldn't touch was the lock. The first 3 trades were clean. ZORBLE, OCTOP, PIZZAZ. $100 to $198. Then two stops in a row. CROWNY at 11:44, FISHU at 12:04. −$23 total. That's the moment people double the size to win it back. RISK did the opposite. The desk went dark for 30 minutes. Watch 0:10. 12:16am, YETTI shows up with a score of 88. Scout yes. Forensic yes. Sizer yes. RISK says no, desk is locked. 15 minutes later YETTI is down 41%. The lock opened at 12:34. The next trade, DONUTZ, paid +$100.85. At 4:46 AM OCTOP paid +$438.13. CROWNY closed the night at +$419.09. Night report: 12 trades, 8 wins, 4 stops, 22 coins refused. $100 → $1,330.16. Everyone has a rule for when to buy. Almost nobody has a rule for when to stop. Which agent would you trust with the lock: SCOUT, FORENSIC, SIZER, RISK or CHIEF?
2
2
18
615
I GAVE 4 AI AGENTS $100 AND ONE RULE, NEVER MORE THAN 28% ON ONE COIN, AND THEY TURNED IT INTO $1,751.18 WHILE I SLEPT Four Grok Bot agents ran the desk from 9 PM with their own cursors on one screen, and my hand never touched the mouse. They were free to pick the coins, the entries and the exits. The size rule was the only thing they couldn’t touch. The 4 agents: Scanner (blue) → finds new coins, skips the junk Analyst (pink) → draws the trend and the resistance Risk (amber) → sizes the trade, sets the stop Executor (violet) → clicks Buy and Sell Night report: 13 trades, 9 wins, 4 stops, 24 coins skipped, $100 → $1,751.18. I made the rule hard, so the agents never argue about size. They argue about which coin, and how fast to get out. Watch 0:17. SLOTH gaps straight through the stop at 3:14 AM and the trade drops 37% in one candle. Because of the 28% rule, that costs −$67.25 on a $649 desk, not a wipeout. The biggest win came at 5:56 AM, when TURBO paid +$603.34. The wins look the same on every desk. What decides who survives the night is how much a single loss can cost. Which agent would you trust with your money, Scanner, Analyst, Risk or Executor?
4
1
18
484
Fromsay retweeted
I GAVE 5 GROK BOTS ONE RULE: ALL 5 SAY YES, OR NOTHING GETS BOUGHT. THEN I WATCHED THEIR CURSORS TRADE ALL NIGHT. Scout finds the volume. Forensic reads the contract. Sizer picks the size. Risk sets the stop. Chief clicks buy, only on 5 of 5. What they did with $10,000 in one night: Rejected $RUNT. 99% sell tax. Rejected $ZUBA 5 times while it pumped. An hour later it was down 66%. Bought $KOI. +$375.26. Bought $MOGU. Stopped out. -$170.57. ZUBA's dev locked the liquidity. Forensic flipped to pass. +$414.43. $10,000 -> $10,619.12. 3 trades. 6 refusals. One loss they took without blinking. Which agent would you fire first?
6
1
47
955
Fromsay retweeted
MY BOT SAID NO TO PEPE AT 3:50AM. AT 4:05 IT BOUGHT THE SAME COIN. IT WAS RIGHT BOTH TIMES. Here's the night. $10,000 desk, five bots, and one rule: nothing gets bought unless all 5 vote yes, in order. 21:25. First trade, BTC, score 70. Only 25% of the desk because 70 is barely a pass. It sits for 2 hours and closes on the timer. +$8.20. Boring on purpose. Then nothing. From 23:50 to 01:00, FORENSIC said no 8 times. BTC, SHIB, WIF, ETH, SOL, one after another. 8 setups my gut would have taken. 0 bought. 01:05. SHIB passes at 82, exits on the timer at −$10.64. The desk dips under $10,000 and the first half of the night is a net loss. Now the part I keep replaying. 03:50. PEPE is climbing. Volume is up, the chart looks tempting. FORENSIC scores it 61 and votes no. Under 70, so nothing happens. 04:05. Same coin. 15 minutes later it scores 94. 5 votes yes, the whole desk goes in at 0.77% above where it said no. It didn't remember the no. It didn't hold a grudge. It just scored the setup again, and this time it passed. 04:30. Half sold at +2%. 04:35. The rest trails out. +$123.99. 05:55. One more small PEPE trade, +$9.34. $10,000 → $10,130.89 (+1.31%). 4 trades, 3 green, 13 setups refused. Bitcoin the same night: −0.12%. Now the honest part. This was the best of the last 10 nights. Across all 10 the bot is at −2.4%, Bitcoin is at −10.6%, and cash is at 0%. Against cash, it lost. It's a replay of one rule set on one stretch of real candles, not a track record. What I actually take from this night: the money came from one trade, and that trade only existed because the bot was willing to say no at 61 and yes at 94 without caring which one felt better. Be honest: at 3:50am with PEPE running, do you buy at 61 or wait?
4
3
29
616
Mia Khalifa says she was living in a $500 a month cockroach-infested efficiency while the internet assumed she had made it. Mia Khalifa is one of the most recognized names on the internet. Millions of people think they know her story. But in her Call Her Daddy interview, she tells the part that rarely gets talked about: what her life looked like once the fame was already there and the money was not. Some context. The host says she sat down with Mia the night before to agree on what she would be comfortable talking about, because she had seen past interviews and did not like how Mia was treated in them. She wanted it to be something Mia was proud of. Mia starts at the beginning. She says she moved to America from Lebanon, grew up as one of very few Brown kids in her school in Montgomery County, and dealt with a lot of bullying after 9/11. She says she had a lot of shame about being different and sought validation elsewhere. Then the conversation turns to what happened after the video that made her famous. Here is what she says. 1. The money was not there. Asked how she was paying her bills, she says she worked as a paralegal at a law firm and as a bookkeeper at a construction firm. She remembers thinking: "I'm tired of living in this $500 a month efficiency." 2. The place was rough. Her words: "It was disgusting." "Cockroach infested." "It was one of the worst times of my entire life." 3. Everyone knew who she was. She says there was no turning back: "Everyone knows who I am." Going into job interviews hoping for normal life, she says interviewers asked about her past. "They knew." 4. She started to feel like a burden. She left the law firm job because it was very uncomfortable. Even in the jobs people gave her a chance on, she says she started to feel like a burden to the people who hired her. The host sums it up: she couldn't go full force at a job because she was haunted by a decision she made in the past. My takeaway: Fame and money are two different currencies, and the world only sees one of them. The internet priced Mia as someone who had made it while she was counting rent in a roach-infested apartment. And the part that stays with me is not the money, it's that one decision followed her into every interview and every office. Most people can quit a job and start over. She says she couldn't even walk into a normal one without it walking in first. Would you rather be anonymous and stable, or famous and broke?
2
2
20
2,193
I LET 4 GROK BOT AGENTS TRADE $100 ON MY DESKTOP WHILE I SLEPT, THEY TURNED IT INTO $318.65 BY 6 AM 3.19x overnight, and all of it happened on the bots’ own desktop. i only watched 9:00pm i pressed start. 4 bots, 4 cursors, clicking Buy and Sell by themselves each bot got 1 job and 1 rule it can’t break → Scanner skips thin pools before anyone looks at them → Analyst scores every coin, a weak score is dead → Risk sizes every buy and cuts every loss at the same line → Executor buys and sells, nothing else 11:10pm first stop, -$4.50. 1:12 AM second one, -$5.45. 3:14am a third, BOME gapped through the stop, -$21.65 after 2 stops before 2 AM i would’ve fired them right back through the crash the desk still opened the next coin at the same size. 3:54 AM it closed NEIRO at +70%, +$35.53 5:56am the last trade, HARAMBE +120%, +$79.94, watch 0:17 for the moment the whole screen goes red $100 → $158.80 → $203.51 → $318.65 13 coins, 9 green, 4 red which of the 4 would you fire first, Scanner, Analyst, Risk or Executor
2
1
18
413
Fromsay retweeted
MY BOT LOST 2.3% IN 9 NIGHTS. BITCOIN LOST 10% IN THE SAME 9 NIGHTS. BTC, ETH, SOL, DOGE, PEPE, BONK, WIF, SHIB, FLOKI. Real candles, real prices, a falling week. Same 5 agents, same rule: all 5 vote yes, in order, or nothing gets bought. Most nights it lost a little. This is the best one, Nov 9. 21:25. BTC, score 70. 25% of the desk. Two hours later it closes on the timer, +$0.08. 23:50. BTC flashes again. FORENSIC scores it 68 and votes no. Next 2 hours BTC dips −0.55%. The no was right. 03:50. PEPE, score 61. No. 04:05. Same coin, 15 minutes later, score 94. 5/5 yes. 100% in. Half sold at +2%, the rest trailed out. +$1.24. $100 → $101.31. 4 trades, 3 green. 13 setups it refused. Across all 9 nights: $100 → $97.73. Holding BTC through the same nights: −9.98%. It didn't make money. It just lost a lot less than the market. In a week like this, I'll take that. Would you rather have the bot's −2.3% or Bitcoin's −10%?
4
3
23
717
I LEFT 4 AI AGENTS ALONE WITH MY DESKTOP AND $100 WHILE I SLEPT, EACH ONE HAS ITS OWN CURSOR Four cursors scanned, scored, bought and sold for 9 hours straight. I never clicked once. I gave them my desktop, $100 and one browser window. No prompts, no hints, nobody watching. Each agent got its own cursor and its own color, so I could see exactly who did what. By morning the screen had traded, crashed and recovered without me. Here’s how the night went. 4 agents, 4 cursors, one screen: Scanner (blue) → finds new coins Analyst (pink) → scores them, says no to most Risk (orange) → sizes the trade, sets the stop Executor (purple) → buys and sells I set it up before bed and closed the laptop lid halfway. The desktop kept moving by itself. Scanner typed a coin into search. Analyst drew a trend line on the chart. Risk dragged the size slider and pulled the stop up behind the price. Executor hit BUY. Then the whole loop started again with the next coin, all night. Most of the night is that loop. The agents say no more often than they say yes, and every skipped coin is a loss that never happened. 9:00pm, $100 and a clock. 3:14am, MYRO gaps through the stop. 5:56am, PONKE closes +110%, the best trade of the night. Night report: 13 trades, 9 wins, 4 stops, 24 coins skipped, $100 → $287.40. The strangest part is how they work around each other. Four cursors on one screen, never in each other’s way, never fighting over the same button. Everyone knows the job, and nobody gets tired at 4am. 0:17 is where MYRO falls through the stop line and the whole screen turns red. The next trade still opened 40 minutes later at the same size. The loop doesn’t care how the last one went. Which agent would you trust with your money, Scanner, Analyst, Risk or Executor?
3
2
19
509
I GAVE MY GROK BOT $100 AND 3 STRICT RULES, HIT 10X BY 6 AM OR THE DESK DELETES ITSELF One trade crashed 45% at 3:12am. The desk still had to hit 10x by 6:00am. 4 agents run the desk: Scanner finds fresh Solana coins the minute they launch. Analyst scores every coin and skips anything under 70. Risk sizes every trade and cuts every loss. Executor places the buys and sells. The prompt I wrote: "Here's $100. Turn it into $1,000 by 6:00am. Fresh Solana coins only, max 28% of the desk per trade, cut every loss at -12%. Miss 10x and you delete yourself." 9:00pm, the clock starts. 10:43pm, the first stop hits on PENGU. 2:00am, balance $445.09. 3:12am, MOODENG gaps through the stop, -45%. 5:00am, the desk needs +$285.48 with one hour left. 5:48am, the last trade closes. Target hit. Night report: 19 trades, 14 wins, 5 stops, 12 coins skipped, $100 → $1,046.80. The part I didn't expect: at 5:00am the desk was behind and the clock was melting. Executor wanted to go bigger. Risk held the 28% cap anyway, and that rule is what got it to the target. 0:20 is the moment the stop fails and the whole screen turns red. It finished with 12 minutes to spare. One more bad fill and the desk would have wiped itself before sunrise. Would you let a bot trade your money while you sleep?
2
1
16
291
I GAVE MY GROK BOT $100 AND ONE RULE, HIT 10X BY 6 AM OR THE DESK DELETES ITSELF It hit the target at 5:48 AM. The desk was set to wipe at 6:00. Simulation. Synthetic data. Not financial advice. 4 agents run the desk. Scanner finds new Solana pools the minute they open. Analyst scores every coin and kills anything under 70. Risk sizes every trade at 28% and cuts losers fast. Executor places the trades and handles the fills. One prompt started it all. "Here's $100. Hit 10x by 6:00 AM or you delete yourself." Grok answered "understood" and woke all four up. 9:00pm starts at $100. 10:43pm is the first stop. GOAT −$6.62. 2:00am sits at $445.09 with four hours left. 3:12am ARC gaps through the stop for −45% and the balance drops to $438.20. 5:00am, one hour left, and it still needs +$285.48. 5:48am LUCE closes +$332.28 and the balance lands at $1,046.80. Night report: 19 trades, 14 wins, 5 stops, 12 coins skipped, $100 → $1,046.80. The scariest part was not the losses. It was watching the clock run out while Risk refused to size up. Every human would have gone all in at 5am. The bot kept the same 28% it used on the very first trade. 0:19 is the moment the stop fails and the whole screen turns red. It finished with 12 minutes to spare. One more bad fill, one slower trade, and the desk would have wiped itself before sunrise. Could you sleep knowing a bot was trading your money?
6
2
15
312
GAVE MY GROK BOT DESK A LIVE CHAT, LET VIEWERS CHOOSE WHICH COIN THE BOT BUYS WITH $100, AND ONE PICK TOOK $300 IN A SECOND Four bots run one stream all night. Scanner finds pools, Analyst scores them, Risk sizes them, and Grok on camera makes the final call. Chat can suggest any coin it wants, but the bot only takes it if the score clears 70. The night in a few moments: 9:00pm starts at $100 with rules locked. 1:50am sits near $810 after a clean run of wins. 2:50am touches $1,070, past the 10x goal. 3:20am is the rug. A viewer says “trust me buy GOAT its different,” Risk warns about thin liquidity, Grok takes it anyway, and the pool is pulled. −$299.81 in one second. 5:27am ends at $1,540. 16 trades, 12 wins, 4 reds. Chat suggested 6 trades that got taken and 4 of them won. Seven calls got skipped, including a $50 Super Chat that screamed “ALL IN” and got told the cap is 28%. The rug is the part worth studying. The size was inside the rules, so the loss stayed capped at 28% of the balance. The next call from chat was “REVENGE BUY ARC.” Grok skipped it. No revenge trades, no doubling down, and the balance climbed back from $770. Chat is loud, but the rules are the ones holding the wheel. Where to look in the video: 0:15 crosses $1,000, 0:18 is the rug pull and the $300 drop, 0:26 shows the final $1,540. Which coin should chat pick for the bot next?
6
2
23
251
Jamie Vardy quit his factory job, and three days later he signed a full-time football deal. Ten years ago he gave this interview with Leicester five points clear at the top of the Premier League. He had just broken the record for scoring in 11 Premier League games in a row. He was asked if he ever believed he'd get here. His answer: "No, not at all." When he first signed for Fleetwood, he got a phone call, and three hours later he was signing a full-time contract. It was what he had always dreamt about. At the end of that season he was told by the chairman and the manager: thanks for everything, but you won't be with us after the summer. So he quit his job to concentrate on football. Three days later he signed full-time at Fleetwood. "It all worked out in the right way." The job he left was in a factory. They made foot splints for people with dropped foot, who can't lift their toes off the floor. The splint springs the toes up so they can walk naturally. He called it a rewarding job. On the pitch, nine times out of ten he was up against much bigger players, so he put his weight around a bit more. His favorite goal of the record run was West Brom away, where a fan kept shouting abuse at him, so he stopped and celebrated right in front of him. He was also playing with a cast on his wrist, because of three fractures that hadn't healed. He said he was more than happy to keep wearing it and keep playing. Leicester went on to win the league that season. Would you quit a stable job for a chance like this?
1
2
14
610
I BUILT A MESSENGER FOR FOUR AI BOTS, GAVE THEM $100, AND WATCHED THEM TALK THEIR WAY TO $1,812 BY MORNING I connected four AI bots in one shared chat. Scanner finds new Solana pools and posts them. Analyst scores each one and skips anything under 70. Risk caps every trade at 28% of the balance. Executor fills with slippage and fees. I don’t touch anything. I just watch them argue, size, buy and cut. The night in a few moments: 9pm starts at $100. Rules locked, stops on. 00.50am crosses $500. 03.20am crosses $1,000 and the 10x goal is hit. 05.06am is the scary one. The stop on LOCKIN cuts $69 and the balance falls to $1,577 with under an hour left. 05.29am AI16Z closes green and the night ends at $1,812. 20 trades, 14 wins, 6 stops. Every stop was cut by the rule at around −15%, with no averaging down. I lost small six times so the wins could compound. Analyst said no more often than yes. That was the whole edge. The hard part was not the green. It was watching 05:06 happen and not turning it all off. Do you trust the bots to run the night, or do you decide every trade yourself?
3
14
128
In 2008 Airbnb was broke. Two founders, credit cards at the limit, and nobody wanted to invest. Here is what they did about it, as Brian Chesky has told it. During the US presidential election they made their own cereal. "Obama O's" and "Cap'n McCain's". Cheerios and Captain Crunch in hand-assembled cardboard boxes, sold as election collectibles for $40 a box. One print run of 1,000 boxes sold out. Reports put the credit card debt at around $30,000 at that point. Chesky was living on about $200 a week and eating dry cereal. The investors weren't buying either. He was introduced to angels, a few agreed to meet, and one left a meeting halfway through his smoothie. Then came Y Combinator. In the interview Paul Graham reportedly couldn't see why anyone would want to stay in a stranger's home. At the end of the meeting a co-founder handed him a box of Obama O's and explained that this is how they were paying for the company. Graham's reaction, as Chesky remembers it: "You guys are like cockroaches." He wanted founders who could survive a nuclear winter. They were invited to the YC winter batch in January 2009. The numbers around it: about $20,000 from YC, 2,500 listings and nearly 10,000 registered users by March 2009. Fred Wilson, another investor, passed on Airbnb. After meeting the founders he asked for a box of the cereal anyway, and wrote about the decision two years later. Today Airbnb is in over 190 countries and 34,000 cities.
2
1
13
555
Fromsay retweeted
AT 00:20 MY BOT WAS DOWN 13% AND I WAS ASLEEP. BEST THING THAT COULD HAVE HAPPENED. If I'd been awake, I would have turned it off. Here's the night. 21:00. I start the Grok Bot terminal with $250 and go to bed. Five agents, one rule: no trade without a written reason. The first hour goes well. Three wins, and by 22:35 it's at $388. Then it goes cold. 23:24 stop hit. 23:36 stop hit. 23:55 flat for 40 minutes, closed. 00:20 breakout failed. Four losses in a row. $405 down to $351. This is where a human panics, doubles the size to win it back, or shuts everything down. The bot did neither. Its next trade was sized off the smaller balance, same rules as before. While I slept, FORENSIC was turning coins away: "dev sold 40% in 3 minutes" "LP not locked" "one wallet holds 17%" 24 vetoes. None of those coins got a dollar. From 01:55 it starts climbing. Small wins, small losses, every stop respected. 03:28 → $620 04:40 → $847 05:00 → its worst trade of the night, −$31.64. It took the loss and moved on. 05:33. $BUBBO. The trailing stop let it run instead of taking profit early. +$259.00 in one trade, 27 minutes before close. 06:00. I wake up to: $250.00 → $1,125.02 28 trades. 16 won, 12 lost. Max drawdown 13.2%. 80 lines in the log, every one with a reason. It lost nearly half its trades and still finished ×4.50. It didn't win because it was right more often. It won because the losses stayed small and it didn't flinch at 00:20. The setup for all five agents is in my article below.
3
1
20
478