tenzo (@0xTenzo) header
tenzo (@0xTenzo) profile picture

AI + automation = money. showing real setups

Somewhere in NYC 🗽
Joined July 2026
Photo by tenzo (@0xTenzo)Photo by tenzo (@0xTenzo)Photo by tenzo (@0xTenzo)Photo by tenzo (@0xTenzo)Photo by tenzo (@0xTenzo)Photo by tenzo (@0xTenzo)Photo by tenzo (@0xTenzo)Photo by tenzo (@0xTenzo)Photo by tenzo (@0xTenzo)Photo by tenzo (@0xTenzo)
I GAVE GROK MY F*CKING MACBOOK AND WENT TO BED. $110 → $12,950 BY MORNING It had its own cursor. It clicked, typed, opened apps, drew on charts and placed 59 trades. All night I pressed exactly two buttons: Allow and APPROVE. Sunday night, BTC had been stuck between $64K and $65K for two days. The task was easy to say and annoying to solve: find a way to make money while the market goes nowhere. Grok spun up 10 agents, each with its own strategy, and put every one through 4 tests. Does it make money. Does it hold up. Does it work on data it has never seen. Does it still hold up on that new data. Fail one and you’re out. Nine were out. The survivor was Range ping-pong. Then the cursor did the setup by itself. It opened the BTC chart and drew the range, 64,180 to 65,060. Long at the floor, short at the ceiling. 40x, both directions, a stop 1.1% outside the range on each side. Then it clicked Go live. All night it bounced between those two lines. 57 trades, every one a small move, every one on almost the whole balance. By 9:25 AM, $110 had turned into $9,658. Most people close the app on a flat day. My bot found 57 trades in one. Then the market opened and broke the range. The stop on a BTC short got hit for −$2,929. That’s exactly where I would have panicked and started revenge trading. The bot didn’t argue with the market. It checked the breakout, decided it was fake, shorted ETH with 95% of the balance, moved the stop to breakeven, and rode ETH from 1,852 down to 1,806. +$6,221. The best trade of the night came five minutes after the worst one. $110 → $12,950. In the same window, BTC moved −0.83%. 40x with 95% of the balance on one trade is as aggressive as it gets. This is one night’s log, not a setup to copy. If you want to run a night like this yourself: - Give it one specific problem, not “make money.” Mine was: the market is flat, find a way anyway - Make strategies compete. Ten in, most should die - Always test on data the strategy has never seen. That’s where fake edges die - Have it draw its levels on the chart, so you can see exactly what it’s trading - Set the stop before it goes live, not after the first loss And the task, roughly as I gave it: You have my Mac. BTC has been stuck in a range for two days. Find a way to make money while it goes nowhere. Run 10 strategies through 4 tests: performance, robustness, unseen data, robustness on unseen data. Kill anything that fails one. Set up the survivor on BTC, ETH and SOL perps and go live. Save this for your next flat night. Two buttons. Allow before bed, APPROVE in the morning. Everything in between was the cursor.
20
3
34
600
tenzo retweeted
YESTERDAY’S WINNING STRATEGY GOT FIRED. MY GROK BOT FLIPPED LONG: $150 → $15,855 OVERNIGHT Have you ever seen a bot kill its own best strategy the day after it worked? Saturday, 8 PM in New York. BTC was down 14% in three days and sitting around $62K. I gave my Grok bot a second night shift, same job as yesterday: come up with 10 perp strategies for BTC, ETH and SOL, score each one on 4 criteria, run whatever survives on $150 at 40x, wake me at 10. And the first thing it did was put yesterday’s winner on trial. - Asia-open momentum, Sharpe 5.9, too clean to be real, cut on sight - Keep shorting the dump, the strategy that won last night. The numbers said today it was just luck. Cut - Funding reset, out - Session-low sweep, out - Range-high breakout and SOL beta chase, both made the final round, both cut - Capitulation reversal, the only one left standing The idea behind the survivor: longs just got flushed, so buy the reclaim and ride the short squeeze. Longs only. It went live at 8:44 PM. BTC had printed $62,223 four minutes earlier. Yesterday it was shorting everything. Tonight it was long a few minutes off the bottom, with zero ego about what worked last night. I’d still have been shorting out of habit. By 10 AM: $150 → $15,855. 31 trades, all longs, 22 green. SOL did most of the work, +$7.7K, and the best single trade was a SOL long for +$3,453 in the middle of the night. Two morning stops cost $342 and $246, and the worst drawdown was 13.7%. 40x with about two thirds of the balance in play is about as aggressive as it gets. This is one night’s log, not a setup to copy. If you want to run something like this yourself, here’s what actually matters: - Make it generate many strategies, not one. Ten ideas, and most of them should die - Kill anything that looks too good. A Sharpe of 6 on paper is usually overfit - Re-run the tryouts every night. Yesterday’s winner can be today’s loser - Let only the survivor trade, and stop it hard when it’s wrong - Run it on paper before you put 40x on anything Here’s the task, roughly as I gave it: Come up with 10 perp strategies for BTC, ETH and SOL. Score each one on 4 criteria. Kill anything that fails. Run the one that survives live on $150. Wake me up at 10 AM. Save it. You’ll want it the next time the market flips on you. Two nights, two opposite strategies, one bot. It doesn’t care which way the market goes, as long as it gets to choose.
22
2
43
973
YESTERDAY’S WINNING STRATEGY GOT FIRED. MY GROK BOT FLIPPED LONG: $150 → $15,855 OVERNIGHT Have you ever seen a bot kill its own best strategy the day after it worked? Saturday, 8 PM in New York. BTC was down 14% in three days and sitting around $62K. I gave my Grok bot a second night shift, same job as yesterday: come up with 10 perp strategies for BTC, ETH and SOL, score each one on 4 criteria, run whatever survives on $150 at 40x, wake me at 10. And the first thing it did was put yesterday’s winner on trial. - Asia-open momentum, Sharpe 5.9, too clean to be real, cut on sight - Keep shorting the dump, the strategy that won last night. The numbers said today it was just luck. Cut - Funding reset, out - Session-low sweep, out - Range-high breakout and SOL beta chase, both made the final round, both cut - Capitulation reversal, the only one left standing The idea behind the survivor: longs just got flushed, so buy the reclaim and ride the short squeeze. Longs only. It went live at 8:44 PM. BTC had printed $62,223 four minutes earlier. Yesterday it was shorting everything. Tonight it was long a few minutes off the bottom, with zero ego about what worked last night. I’d still have been shorting out of habit. By 10 AM: $150 → $15,855. 31 trades, all longs, 22 green. SOL did most of the work, +$7.7K, and the best single trade was a SOL long for +$3,453 in the middle of the night. Two morning stops cost $342 and $246, and the worst drawdown was 13.7%. 40x with about two thirds of the balance in play is about as aggressive as it gets. This is one night’s log, not a setup to copy. If you want to run something like this yourself, here’s what actually matters: - Make it generate many strategies, not one. Ten ideas, and most of them should die - Kill anything that looks too good. A Sharpe of 6 on paper is usually overfit - Re-run the tryouts every night. Yesterday’s winner can be today’s loser - Let only the survivor trade, and stop it hard when it’s wrong - Run it on paper before you put 40x on anything Here’s the task, roughly as I gave it: Come up with 10 perp strategies for BTC, ETH and SOL. Score each one on 4 criteria. Kill anything that fails. Run the one that survives live on $150. Wake me up at 10 AM. Save it. You’ll want it the next time the market flips on you. Two nights, two opposite strategies, one bot. It doesn’t care which way the market goes, as long as it gets to choose.
22
2
43
973
A DEV TRIED TO RUG ME. HE WAS 0.4 SECONDS TOO SLOW -97.6% for everyone still holding. -3% for me. i didn’t even see it happen i put in $101 and gave memecoin trading to a swarm of 7 Grok Bot agents for 24 hours. one prompt, and the swarm went to work the STONK trade. we were already in the position when Guard saw the dev start pulling liquidity. “ALERT”. Core called “exit 100%”. Closer sold in 0.4 seconds we got out at -3%. a minute later the coin was down 97.6% each agent does its own job, they decide together: 01 / Scout catches new launches github.com/swcstudiospace/Pu… 02 / Whale watches smart wallets github.com/swcstudiospace/Pu… 03 / Guard checks for rugs before the buy and keeps watching after github.com/swcstudiospace/Pu… 04 / Edge separates hype from the real price github.com/swcstudiospace/Pu… 05 / Risk max 2% risk per trade github.com/swcstudiospace/Pu… 06 / Closer takes the profit, or gets you out github.com/swcstudiospace/Pu… 07 / Grok Core one word: BUY or SKIP github.com/swcstudiospace/Pu… the swarm also learns. after the first rugs it wrote itself a rule: “dev holds over 25% → auto skip”. after STONK, a second one: “liquidity getting unlocked → exit instantly” github.com/swcstudiospace/Pu… every coin it skipped died. SOLDOG -99.2%, HODLR -100%, FRENZ -99.9%, GIGAPEPE -94.6%. the ones it bought made all the profit. best: LEEROY +212% $101 → $3,199 in 24 hours. 152 trades, 78% green. 214 rugs dodged, 0 rugs bought, 1 rug survived at -3% protection doesn’t stop at the entry. the most dangerous moment is when you’re already in a bot that writes its own rules gets better every day 0.4 seconds vs a human checking their phone. that’s the difference between -3% and -97% the dev was fast. the swarm was faster save this the full setup and prompts are below 👇
$500 → $11,101 in 5 days. my Grok memecoin desk got a 7th seat: Jev. it can't talk. it answers 4 closed questions with probabilities, and nothing gets bought below the gate. full setup below. save it.
17
4
50
1,036
tenzo retweeted
HACKERS PROBABLY ALREADY HAVE YOUR EMAIL AND PASSWORD. I CHECKED MINE: 17 LEAKS every time a site you signed up for gets hacked, your email and password end up in a database anyone can search. that's how accounts get taken over. so i sent 4 Grok Bots to check everything tied to my email. 4 minutes later: → my email was in 17 data breaches → 6 of them leaked a password too → 2 of those passwords still opened real accounts → my phone number showed up in 9 of the leaks → my home address was for sale on 34 people-search sites → 21 accounts i forgot i ever made, 4 still had my card saved the worst one: a 2019 leak with my password in plain text. i was still using it for Amazon. what they fixed: → changed every leaked password, moved Amazon to a passkey with my Touch ID → deleted the 21 dead accounts and removed my card from all 4 → sent removal requests to all 34 people-search sites → set alerts for any new leak with my email or phone watch 0:13, that's my home address getting wiped off the internet the prompt: check my email and phone in every public data breach, tell me which leaked passwords still work, find every site selling my address and every account i forgot, then fix all of it and show me the list save this and check yours tonight
9
3
22
961
MY GROK BOT KILLED 9 OF ITS OWN STRATEGIES. THE 10TH TURNED $200 INTO $13,920 OVERNIGHT Have you ever watched a bot throw out a Sharpe 6.4 because it looked too good to be true? Friday, 9 PM in New York. BTC had just lost $70K and kept sliding. I gave my Grok bot a job for the night: come up with 10 strategies for BTC, ETH and SOL perps, score every one on 4 criteria, trash whatever fails, run the survivor on $200. Wake me at noon. Then I went to sleep. It spent the first 20 minutes trying to kill its own ideas. Here’s how they went: - NY-open momentum, out first. Sharpe 6.4. Real edges almost never look that clean - Buy the dip, out at p = 31%. That result could easily be luck, and in a falling market it’s just catching knives - Funding flip, out - VWAP reclaim, made the final round, still cut - BTC to SOL lead-lag, made the final round too, still cut - Liquidation cascade short, the only one left standing That’s the part I keep thinking about. I would have fallen in love with the Sharpe 6.4 and run it at 9:01. The survivor is simple: when longs start getting liquidated in a chain, short the break and cover on the first bounce. Live at 9:20 PM. $200, 40x, shorts only. The first cascade hit ten minutes later. $537 by midnight. Overnight the market chopped sideways, the bot took small wins, ate a few stops, and sat at $1,083 by 7 AM. Then at 8:15 AM BTC broke again, from $67,150 down to $64,770, and the cascade it had been waiting for all night finally came. The best trade was a BTC short from 8:55 to 10:35 for +$2,699, and ETH and SOL shorts stacked on top, several of them over $1,000 each. It wasn’t clean. A BTC stop cost $252, and the last ETH short at 11:35 got knocked out by a bounce for −$1,468, off a peak of $15,388. 32 trades, all shorts, 26 green, max drawdown 16%. 40x with about three quarters of the balance in play is as aggressive as it gets. This is a log of one night, not a setup to copy. What I’m stealing from this, even without a bot: - If a backtest looks too good, assume it’s overfit until it proves otherwise - A high p-value means luck, no matter how pretty the chart is - Make every idea fight for its life. Most of them should die - Trade the market you’re in. In a falling market the dip buyer died and the short survived Here’s the task, roughly as I gave it: Come up with 10 perp strategies for BTC, ETH and SOL. Score each one on 4 criteria. Trash anything that fails. Run whatever survives live on $200. Wake me up at noon. Save it for the next red night. At noon the report came in. I hit APPROVE. The night shift keeps going.
21
4
44
1,411
tenzo retweeted
I LET GROK BOT PICK THE WINNER $500 at 2:15 PM. $8,594 at 12:15 AM a marble race of 10 real Solana memecoins, each one rolling on its own real 15-minute candles. the winner started as the second smallest bet on the board. the grid: > RONIN, $200, the favorite > MORI, $175 > greg, $75 > Shitcoin, $50 the race: > RONIN jumps out to +68%. Grok cashes it out and moves everything onto MORI > MORI runs to +256%. a quarter comes off and goes onto HWAN, which does +111% > Shitcoin never leaves the start. cut at -9% > MORI hits its top at +332%. Grok sells and puts $2,040, almost the whole stack, on greg > the very next candle, greg does +70% > late at night, 10% of greg comes off at +254% kicked off the track before the start as rug risk: > POOR, later -81% > CSI, later -80% > Rex the finish: greg crosses the line first, +361%. one close, $5,682. $500 to $8,594 in 10 hours. the honest part: $2,040 on one marble was almost the whole stack. if greg stalls, this is a post about losing everything. Grok never bet on who would win. it kept moving the money onto whoever was winning, and about four hours in, that turned out to be a $75 marble called greg.
18
4
37
1,025
tenzo retweeted
i pasted one prompt into Grok Bot & it showed me something most traders never see on Hyperliquid every position is public. entry. size. leverage. & the EXACT price where each whale gets liquidated. one API call. no key. no login. so i asked Grok Bot to stack all of it on one chart. here’s what it found: $64M of whale longs sitting just 2% under the price that’s not support. that’s a row of dominoes. here’s how a cascade works: → price dips into the first liquidation → the exchange force-sells that whale’s position → the forced selling pushes price into the next one → which force-sells the next whale → & the next nobody decides to dump. the liquidations do it on their own. what the bot does with it: 1. MAP THE ZONE tracks the top 200 wallets every 30 seconds & marks every cluster of liquidations close to price 2. RIDE THE CASCADE the moment the first one actually prints, it opens a small short. not before 3. COVER EARLY it closes before the zone is fully cleared. the last domino is where everyone else gets greedy 4. BUY THE FLUSH when the wick overshoots & funding flips, it buys the bounce. small the rules that keep it alive: - max 3x, so its own liquidation price is never on anyone’s map - $10 per trade, auto-stop at -$30 a day - 30-day backtest, 300 paper signals, live only when you type “go live” the whales can’t hide their stops. you just need something watching 24/7. comment “PROMPT” & i’ll send you the full copy-paste version 👇
Grok Bot builds MOST powerful trading agents right now It gives you an army of agents that can do work for you 24/7 If you set it up correctly, you gain super powers In this article, I cover how to build a team of 24/7 AI trading agents from scratch
3
1
11
370
tenzo retweeted
SOMEONE DRAINED $11,400 FROM MY WALLET AT 2 AM. THEY DIDN'T KNOW MY GROK BOT WAS AWAKE І was asleep the whole time. 3 weeks ago i signed an "airdrop claim" on a fake site. that one signature gave a drainer unlimited access to my USDC. it waited, then took everything at 2:07 AM. before i even woke up, Grok Bot: → traced the money through 6 wallets and a swap to USDT → found $9,800 landing on an exchange → sent the exchange's compliance team the full trail at 2:33 → got the account frozen at 3:12 → filed an FBI ic3 complaint and a police report → revoked 23 forgotten token approvals → moved what was left to a fresh wallet and blocked the drainer at 8:41 AM the $9,800 was back in my wallet. $1,600 went through a mixer and it's gone for good. watch 0:21, that's the money coming back save this so it never happens to you
9
4
5
26
1,003
MY GROK BOT REFUSED 7 COINS. ALL 7 F*CKING RUGGED and the ones it did buy turned $25 into $1,475 in 24 hours i put in $25 and handed memecoin trading to a swarm of 7 Grok Bot agents. didn’t touch anything for a full day every agent has its own job, all of them think with one brain: 01 / Scout catches fresh launches and watches the volume github.com/swcstudiospace/Pu… 02 / Whale checks if smart wallets got in github.com/swcstudiospace/Pu… 03 / Guard rug check: mint, liquidity, dev wallet, honeypot github.com/swcstudiospace/Pu… 04 / Edge compares the real price action with the hype github.com/swcstudiospace/Pu… 05 / Risk sizes every trade so one loss costs 2% max, stop on every entry github.com/swcstudiospace/Pu… 06 / Closer takes the profit github.com/swcstudiospace/Pu… 07 / Grok Core hears everyone out and gives one word: BUY or SKIP github.com/swcstudiospace/Pu… the best part wasn’t what it bought. it was what it skipped: $SOLDOG dev held 31%. bot said no. 33 minutes later: -99.2% $HODLR mint still on. 10 minutes later: -100% $FRENZ honeypot, couldn’t sell. 13 minutes later: -99.9% $GIGAPEPE top 10 wallets held 64%. 35 minutes later: -94.6% i tracked 7 coins it skipped. all 7 died. every time Guard dropped one line in the chat: “called it” the money came from $FLOAT, $AMD, $HIGH, $ZELDA and $MAMESUKE. best trade: $LEEROY, +212% on one close $25 → $1,475 in 24 hours. 152 trades, 78% green. 980 coins skipped, 214 rugs dodged, 0 rugs bought. 41 ms per decision in memecoins the money isn’t in buying, it’s in saying no. any of those 7 would have taken everything one bot falls for hype easily. when 7 agents check each other, a rug doesn’t get through small size, hard stop. the swarm works, i sleep i put in $25, a day later it was $1,475, and every rug it skipped went to dust save this the full setup and prompts are below 👇
$500 → $11,101 in 5 days. my Grok memecoin desk got a 7th seat: Jev. it can't talk. it answers 4 closed questions with probabilities, and nothing gets bought below the gate. full setup below. save it.
18
4
1
34
694
10 TRADES. 141X. WHAT THE F*CK. $49 → $6,920 ON SOLANA MEMECOINS Have you ever seen a Grok bot’s whole trading day fit on one screen? 96 dots, one for every 15-minute scan. This is the most minimal thing Grok has built for me yet. I put in $49 in USDC on Solana, pasted one prompt, and it ran every new pool through 6 gates: liquidity, holders, top 10 share, volume, FDV, fresh buyers. Fail one and it’s gone. Clear all six and already moving, it buys. Over 2,200 coins came through in 24 hours. It bought 10. That’s less than half a percent. I buy 10 coins before breakfast. One of the ten was AI. Last week the bot skipped AI twice and watched it dump. This time it cleared all six gates, and the bot was in and out in 15 minutes. The trades that made the day: - AI, +153%, in and out in 15 minutes - FOOD, +1,743%, the big one, held almost five hours - Punch, +220%, the late-night one, 11:15 PM to 1:45 AM FOOD is why the day looks the way it does. All six gates green, volume climbing, so it went in big and rode it from 4:30 to 9:15 PM. Two of the ten lost. That’s the whole red side of the day. The exact prompt: Run Grok Bot in Filter Mode for the next 24 hours. Push every new Solana pool through 6 gates. Gates: liquidity, holders, top-10 share, volume, FDV and fresh buyers. Drop a coin at the first red flag. Buy only what clears all 6 and is already moving. Size up only on the strongest setups. Take profit in pieces, cut losers fast, reinvest everything. Keep it minimal: show my whole day as 96 dots, one for every 15-minute scan. Save it. Five lines, 96 dots. Most of those dots were the bot saying no. That’s the whole strategy.
20
3
47
3,141
MY GROK BOT JUST CAUSED A G4 STORM. $98 → $8,472 IN 24 HOURS Have you ever watched your balance turn into northern lights? Every win makes the aurora brighter. I pasted one prompt into Grok Bot, turned on Swipe Mode for a day, and it dealt me every new Solana pool as a card. Red flag, swipe left. Clean and already moving, swipe right. And up top, the aurora: every new balance level is a bigger storm. G1 at $500, G2 at $1,000, G3 at $2,500, G4 at $5,000. The first thing it did was swipe left on FAIR and pants. FAIR paid out on this bot a few days ago. This time it showed red flags, and it fell 60% that day. pants fell 81%. Then it tagged Max as an X-class setup, sized up, and was in and out in 15 minutes at +674%. The first “It’s a Pump” of the day. I can’t pick what to eat in 15 minutes. The trades that lit up the sky: - Max, +674%, X-class, in and out in 15 minutes - FLC, +$3,737, +1,055%, the trade of the day, 10 PM to 12:45 AM - paws, +154%, an old friend, held for over twelve hours FLC alone was almost half the day’s profit. One trade, under three hours. And paws just sat there from 3:30 PM to 3:45 AM before it sold. The moment it closed, the sky went G4 and the balance crossed $5,000. The exact prompt: Run Grok Bot in Swipe Mode for the next 24 hours and deal me every new Solana pool as a card. Swipe left on any red flag: thin liquidity, whales in the top-10, fake volume or no new buyers. Swipe right only on clean cards that are already moving. Size up only on the strongest setups of the day. Take profit in pieces, cut losers fast and reinvest everything. Show my run as northern lights: every win makes the aurora brighter, every new balance level is a bigger storm. Save it. Five lines, and you get your own aurora. Best space weather report I’ve ever read
23
2
70
1,680
80 MILLISECONDS. THAT’S ALL IT TOOK TO CATCH A MEMECOIN DEV WHO CAN STILL MINT I dropped FROGCOIN’s docs into XERJ Reader, a whitepaper, a tokenomics file and an FAQ, and asked the first question you should ask about any memecoin: can the dev mint more tokens? 80 ms later it pointed at the exact line. “Mint authority is retained by the deployer wallet.” Whitepaper, page 4, line 12. I sent the same question over the same files to Claude Haiku 5.5 and Claude Opus 5.5. Both got it right too. They just had to read everything first. Opus started answering after about 5 seconds, Haiku after about 7. Then I checked what one answer costs, scaled to a million questions: - XERJ Reader, $6.50 - Claude Haiku 5.5, $172 - Claude Opus 5.5, $6,880 On this run that’s over 1,000 times cheaper than Opus. Second question: is liquidity locked? Reader went through all three files and came back with NO ANSWER, not in the documents. For a memecoin, that is the answer. If the team never says the liquidity is locked anywhere, I’m not buying. That’s what I like about it. For a rug check you don’t need a nice paragraph. You need the exact line and the page number so you can check it yourself. And sometimes “not in the documents” is the most useful thing a tool can tell you. Ask the docs two straight questions before you buy. It either shows you the line or tells you there isn’t one. It doesn’t make anything up. $0.01 per 1M input tokens, output free, 80 ms. Waitlist👇
🤝 Paid partnership
I've built a new AI model that is not allowed to generate text. Instead, it produces a verifiable program that answers your prompt using the original reference. XERJ Reader-1 is a neurosymbolic model for working with data. It points to the exact words in your document, writes a small program over them, and exact code computes the answer: • 20-90x faster • 10-1,000x cheaper (output tokens free) • every answer is a program you can replay The shortest path to automating data work.
13
43
1,019
I WENT TO BED UNEMPLOYED AND WOKE UP WITH 3 F*CKING INTERVIEWS my Grok Bot went through 1,240 jobs while i slept job hunting used to eat 3 hours of my evening, every evening. scroll listings, tweak the resume for each one, write a cover letter, hit send, hear nothing back. so i gave it to the night shift lights out at 11:30 PM. 6 agents worked all night, one lead agent handing out the work. 1,549 tasks done by morning 01 / Scout went through 1,240 jobs on LinkedIn, Indeed, Wellfound, YC Jobs and company sites. threw out the duplicates 02 / Matcher kept only the ones that match me 80% or more. 86 made it. every rejected one got a reason: “needs 8+ years in healthcare”, “office 5 days a week”, “below my salary floor” github.com/elie222/botdirect… 03 / Tailor rewrote my resume for every job. “made screens for a fintech app” became “shipped the card flow used by 2M people”. the numbers were already in my project notes, it just put them up front 04 / Writer a separate cover letter for every job. 120 words, the hiring manager’s name, no “I am writing to apply” github.com/elie222/botdirect… 05 / Applier sent 41 applications, only the best fits. filled in the forms and attached my portfolio itself 06 / Closer answered recruiters the second they wrote back, offered times and put the calls on my calendar github.com/elie222/botdirect… 7 AM, my calendar: Tuesday 10:00 · Kitewave · design chat Wednesday 14:30 · Lumen Labs · portfolio review Thursday 11:00 · Orbitly · hiring manager call 1,240 jobs → 86 matches → 41 applications → 7 replies → 3 interviews i didn’t type a single application or send a single cold email. every send followed my rules: 80%+ match, remote or NYC, $140k+ 40 sharp applications beat 100 blind ones. the “under 80%, don’t send” rule does all the work a resume written for the job beats one resume for everything. you don’t have time for that by hand. the bot does recruiters book faster when you answer right away, even at 3 AM i went to bed without a job and woke up with 3 interviews on my calendar save this the full setup and prompts are below 👇
Grok Bot builds MOST powerful trading agents right now It gives you an army of agents that can do work for you 24/7 If you set it up correctly, you gain super powers In this article, I cover how to build a team of 24/7 AI trading agents from scratch
11
4
43
1,004
MY GROK BOT TURNED MEMECOINS INTO A VIDEO GAME. THEN IT BEAT IT. $97 → $4,952 IN 24 HOURS Combo x7, LV 4, and one Legendary card it held for 12 straight hours. This time I asked for an arcade. Every 15 minutes a new Solana pool gets dealt as a card, and every card goes through six checks. Any red flag, swipe left. Clean and already moving, swipe right. Every win in a row builds the combo, and every time the wallet grows, you level up. LV 4 unlocked at “Balance hit $2,500.” I’ve played games with worse progression systems. The red flags were loud again. HCAT had $8K of liquidity, KCAT had 92% sitting in the top 10 wallets. Both got swiped left, both rugged that day, −94% and −73%. The Legendary card was BOXCAT: 1,341 holders, all six checks green. It went in at 1 PM and held for 12 hours before closing at +386%. Twelve hours. I can’t hold a memecoin through lunch. The matches that paid the most: - NIHAO, +$777, the biggest close of the day, at 5 AM - MAMESUKE, +$666, skipped last week, back as a winner - GATO, +$605, an old regular - ClapCat, +$457, part of the early-morning run - BOXCAT, +$448, the Legendary, held 12 hours Three bad dates, STONK, PAID and DESSA, cut between −8% and −14%, about $107 all together. DESSA keeps losing on this bot, and the bot keeps not caring. The pumps sped up the wallet, and the night trades on the bigger balance did the rest. The exact prompt: Run Grok Bot in Swipe Mode like an arcade game for the next 24 hours. Deal me every new Solana pool as a card. Swipe left on any red flag: thin liquidity, whales in the top-10, fake volume or no new buyers. Swipe right only on clean cards that are already moving. Size up only on the legendary setup of the day. Take profit in pieces, cut losers fast and reinvest everything. Show my combo streak and level me up every time the wallet grows. Save it. Four lines, and this time it comes with a level system. Game over. High score: $4,952.
19
1
49
2,344