Joying.eth

all in on memecoins // prediction markets // ex Bet365

[Joying.eth @Joyingeth]

https://t.co/F3omRjvSpl

[Joying.eth @Joyingeth]

@0xNeuroBlue great github thanks

[Joying.eth @Joyingeth]

Grok Bot blocked my $500 deposit and cut me to $100

$100 → $5,734.62. 57X. One night flat

11:32 PM. Grabbed 3 traders off the screen and fed Grok Bot the rules:

> mirror these 3 traders
> exit when they exit
> shut it all down by 7am
> starting with $500 tonight

Two lines came back:

"go with $100. if it works, scale up tomorrow."
"first night with these 3. I don't have data on them yet."

Typed "fine". Loaded the app, keyed in 500, wiped it, keyed in 100. Face ID, confirmed.

$400 sat on my card.

12 trades mirrored. 9 green. 3 red: −$11.52, −$27.60 and −$79.20.

3:14 AM. $2.28 left in cash. Everything else was in.

Best hit: $330 → $2,442. +640%. 3:57 AM. Trader sold, bot sold, same tick.

The 7:00 AM lock sealed the last one at +$942.04.

One trader carried the run: +$3,215.32. The other two landed +$1,885.00 and +$534.30.

$100 → $5,734.62.
+$5,634.62. +5,634%.

Pulled at 7:03 AM.

Its morning ping: "it worked. you can scale up tomorrow."

The only bot I'd trust with more is the one that told me to start with less.

[Joying.eth @Joyingeth]

https://t.co/F3omRjvSpl

[Joying.eth @Joyingeth]

I STILL CAN'T PROCESS THIS, MY GROK BOTS TRADED 5 COINS AT ONCE WHILE I WAS OUT COLD, 4 COINS BLED THROUGH THE NIGHT AND THEY STILL TURNED MY $8,506 INTO $17,600

Grok Bot landed Coinbase, Robinhood and eToro connectors last week. it places trades on its own now

at a hedge fund each trader gets one book. my bots worked 5 desks, 5 separate teams, all under one rulebook

not a single coin gets funded until it clears 6 rules:
→ top 10 wallets hold under 60%
→ pool age over 1 hour
→ never buy mid-rocket
→ liquidity above $12K
→ 2 consecutive losses and the bot steps out
→ an X post counts as research, never the trigger

the night:
→ WEFT crashed 51%. my WEFT desk pulled +152%
→ startupSI sank 42%. that desk pulled +156%
→ FTR dropped 23%. that desk made +30%
→ UHI fell 13%. that desk made +335%
→ GTA6 was the only green one at +14%. that desk did +46%

10% of every desk sat in the coin overnight, the rest played the bounces
→ 8:00pm UHI desk enters at $359K, exits at $914K by 9:30pm, +$1,261
→ 3:10am re-enters at $278K, out at $407K, +$1,286

22 winners, 14 losers, no loss held past -8%

the floor splits into 5 offices. Coinbase, Robinhood, eToro, pumpfun and fomo. UHI turned out to be the strongest desk

i didn't make a single trade all night. the rules ran everything, an Elon post would have been my first real entry

would you trust a bot to trade your Coinbase account overnight?

[Joying.eth @Joyingeth]

https://t.co/F3omRjvSpl

[Mojunel.eth @mojunel]

Broke the cost floor with a team

$8 brain plans. $0.50 hands type.

I wired four agent files and each one carries its own model.

Fable is the brain.
Plans the work, splits it, judges what comes back.
Runs only when I type /brain.

Opus 5.5 takes the builder role. Builds from the brief, nothing else.

Sonnet 5.5 is the checker. Reviews the output. Edits nothing.

Haiku 5.5 runs the repetitive tasks and never touches code.

80x between the brain and the runner.

The brain delegates three kinds of work:

1. Independent parts
2. Work too big for one session
3. Repeatable tasks

One brain plans. Three hands deliver.

I describe the job
→ Fable writes the spec at $8 / $40
→ Opus builds at $4 / $20
→ Sonnet reviews at $2 / $10
→ Haiku runs routine at $0.10 / $0.50
→ Builder fails twice, Fable takes over

Never two helpers on the same file at once. Checker signs off before done.

One CLAUDE.md line: consult the advisor before the approach, on a repeating error, before calling done.

Agent teams in plan mode burn 7x the tokens of a solo session. Worth it only when the job outgrows one sitting.

Four files and one line

[Joying.eth @Joyingeth]

I DROPPED 4 FREE GITHUB REPOS INTO GROK BOT. OVERNIGHT THEY FLIPPED MY $231.70 INTO $3,892.78

16.80x across 8 hours 38 minutes, then i cashed everything out

9:58pm, my only input under the links: "add these repos to our setup"

every repo turned into its own agent:

01 / Scanner = GMGN Skills, filters thin pools
https://t.co/QGuUo3fw8r

02 / Analyst = TradingAgents, 121K stars, bull vs bear debate before each buy
https://t.co/IX5yJmcAf3

03 / Executor = Jupiter Agent Skills, handles the swap
https://t.co/6Wxhdh9JDy

04 / Brain = xAI SDK, sells, -6% stop on every buy
https://t.co/ltD0BDd7OW

11 sells, 9 green, 2 red, -$17.58 across the reds, both caught by the -6% stop

best sell: +$1,049.96 at 3:50am

$231.70 → $1,069.28 → $3,892.78

i used to think GitHub was a developer thing. i only dropped links

which one would you switch out?

all 4 are free. keep this for tonight

[6AWORON0k @6AW0RON0k]

https://t.co/3QYDZOhKuW

[Joying.eth @Joyingeth]

I tested an AI app that measures piles of gravel, sand, and dirt just by walking around them with a phone camera.

we checked our storage yard first. it was completely full. the numbers weren't.

$14,000 in material vanished from our inventory count. our site manager kept losing track of deliveries no matter how many times he recounted manually. we put the phone on record and the app calculated the exact volume in 30 seconds.

on my iPhone 15 Pro it:
scanned the entire gravel pile
calculated 659 yd³ volume instantly
estimated 922 tons total weight with high surface confidence

all going to the processing server:
IP → 142.250.***.89
location → Oregon, US
provider → AWS Cloud
app version → 4.2 Pro
scanner → LiDAR 3D Mesh
last sync from my phone → 2 min ago

it was still updating elevation maps while i was walking around it.

how it works: point your phone at the base, walk in a circle around the pile, keep the camera steady. the app builds a 3D point cloud and computes volume and tonnage automatically.

no more climbing massive unstable piles with measuring tapes.

changed how we track inventory from our office. not from the site.

the full step by step is below. takes 7 minutes. save it so you don't lose it and send it to anyone who works in construction or landscaping.

[Ridark @ridark_eth]

https://t.co/cVH81pBj6h

[meya. @birmeya]

I still can't believe what my pc just pulled off... I gave Gemini and Claude a single prompt, and it generated a complete, fully rigged 3D game character with custom animations from scratch.

Not just a static picture. A full turnaround sheet, exported straight into a 3D engine as a "Mushroom King" ready for a Roblox game, complete with walking, running, and special attack states.

It happened in under a minute while I was sitting at my desk testing workflow loops.

My project partner's design concept, text prompts handled by the AI, organic sculpted shapes, low-poly cartoon style, and exact keyframe animations for idle, jump, and combat, all generated seamlessly in one go.

I started clicking through the settings panel. Next to the character model, I saw the full animation tree: Cap Slam, Spore Cloud, Mushroom Roll. Every single action fully parameterized and ready to import.

No tedious 3D modeling for weeks in Blender. No manual rigging or weight painting. Just one prompt, and the AI built a production-ready game asset with active physics and states.

I saved the project files and closed the editor. Who needs a whole design team for rapid prototyping anymore, I don't know. Tomorrow I have to show this to the team.

Any game dev pipeline you're currently running can do the same. If you want to test yours, open up Gemini alongside your design agent, feed it a detailed character concept prompt, and watch it generate complete 3D assets in seconds.

[6AWORON0k @6AW0RON0k]

$163.85 TO BED. $17,048.92 BY MORNING.
GROK BOT CLONED 4 TRADERS 1 SECOND BEHIND AND DROPPED ONE AT 1:53 AM

104X. ONE NIGHT.

11:35 PM. I loaded $163.85 from my card into the app. Face ID, done.

I never chose a single coin. I picked 4 traders off the leaderboard and let them handle the rest.

Then I sent Grok Bot one message before sleep:

> copy 4 traders
> buy 1s after they buy
> sell when they sell
> take profit at 10x
> lock everything by 7am

The entire strategy. 5 lines.

It never trades on its own. 4 agents move every coin down the chain:

> Watcher spots the trader buy
> Copier fires 1s later
> Executor fills the order
> Risk handles the exits

Around 1:50 AM one trader started losing. 2 back-to-back reds: −$10.03 and −$11.04.

At 1:53 AM Risk cut him off. Nothing from me. No manual input.

He kept buying. The bot kept watching and never followed.

His next 2 buys dumped right after: −38% and −27%. Grok Bot sat out of both.

Those 2 reds were the only losing trades all night. 11 copied. 9 green.

One trade ran $58 to $580. The rule closed it at exactly 10x while the trader was still holding.

At 2:00 AM I had $4.02 left in cash. Everything else was live.

At 7:00 AM the bot locked every open position. That lock alone added +$10,567.94.

Then this message was waiting:

[Neuro @0xNeuroBlue]

300 agents tearing apart one app, all at once.

reverse engineering usually looks the same. a single model crawling through the whole app, forcing the full architecture into one context window.

REA breaks that.

it carves the app into isolated targets you can probe on their own - functions, dependencies, call paths, state transitions, UI logic.

K3 then throws 300 agents at that map in parallel.

the old path:

single model → full codebase → one long reasoning chain

the new one:

one app → hundreds of focused probes → one reassembled system

the swarm pieces the picture together.

that changes reverse engineering. instead of one model wrestling the full codebase, 300 agents divide and conquer it.

[Joying.eth @Joyingeth]

NOT ON THE APP STORE. GROK BOT MADE THIS, I FELL ASLEEP, IT TURNED $347.86 INTO $3,048.56

from 11:47pm to 6:36am. 8.76x, 13 trades, i touched none of them

i told Grok Bot to make me a terminal to run off my phone. it put this one together, i dropped it in my Trading folder, tapped Run and went to sleep

4 agents inside
Scanner tosses anything short on liquidity
Analyst scores whatever gets through
Executor buys
Risk attaches a -6% stop to each buy

2:35am i opened my eyes and checked the app. same balance, positions firing open and shut on their own. went back to sleep

out of 13 sells, 10 hit green and 3 went red. those 3 lost $18.33 combined

top trade was a coin named scamsem, that name. +$566.46 in 46 minutes
and Diamond Hands, held true to its name, 1:27am to 5:00am, closed at 5.65x and 6.84x

8.76x is solid, but what i actually like is that this app belongs to me. nobody touches the rules except me

would you hand your money to an app that doesn't live on the App Store, or is that insane?

[Joying.eth @Joyingeth]

https://t.co/F3omRjvSpl

[meya. @birmeya]

BEYOND BROKEN.
My pipeline graded 9 builds. The 3 top scores still read as flat white toys.

A checklist tracks features: eyes, mouth, silhouette. It never counts how much dark machinery shows through the shell.

Three builds passed every line. All three shipped a clean white diagram.

The fix: crop both to the object. Count dark pixels. Demand 0.8 of the reference.

I burned roughly 430,000 tokens on the moth across 9 rounds before I killed it. That miss wrote the density gate.

1. Measure density before round 1. Hydra scored 0.57. One targeted round pulled it to 0.80. Fly hit 0.96 once panels split.
2. Pull proportions before code. Piranha jaws at 0.52 against 0.37 in the picture. Four fractions in. Landed at 0.35.
3. Two builds at a time, review after round 2. One still runs about 400,000 tokens. Late review kills the budget.

Before, every build ran until the checklist passed.
Now the checklist is where the actual numbers begin.

[6AWORON0k @6AW0RON0k]

I reloaded a $100 desk for nine days
It killed 99.5% of everything that touched it.

$100 to $4,812 across 64 trades.

My funnel ran five agents.
1. Finder scanned 14,200 fresh pairs, dropped 13,888 on liquidity and age
2. Auditor checked contracts and holders, killed 194
3. Judge scored survivors 0 to 100, rejected 54 below 80
4. Stager sized every order at 4% of the book, placed the stop before entry
5. Closer exited in tiers and trailed the rest

Stager and Closer never killed a single coin. One only sizes. The other only sells. For every trade my desk took, about 222 pairs died on the way down.

39 winners, 25 losers. Most profit came from the exit ladder that leaves winners running. Half out at double, a quarter at triple, the rest trailing 30% under the high.

Three red days.
Day 3: -7.9%, two stops back to back, no re-entry.
Day 6: -12.0%, daily stop tripped at -10% and locked the desk until midnight.
Day 8: -12.7%, a coin climbed 180% and the 30% trail bled most of it back.

After day 8 I changed one line. Trail from 30% to 25%. Tested it for a full day against the same kill counts. Day 9 closed at $4,812.

The days with the most refusals paid the best.

Would you trust your desk to refuse 99.5% of what walks in?

[Joying.eth @Joyingeth]

$149 TO $1,947.37 BY MORNING, GROK BOT COPIED TWO WALLETS STACKING $37K WHILE I SLEPT

13.07x. Withdrew all of it. They still have no clue I exist

7:55pm, $149 in through Face ID

Then I followed the top 2 on the 24h leaderboard
1st trader, +$21,447.22 in 24h
2nd trader, +$15,617.46 in 24h

One prompt to Grok Bot
→ copy every buy 30s after them
→ sell when they sell
→ take profit at 10x
→ lock the rest by 6:30am

8:01pm it tailed the 1st trader into a coin at $52,789 market cap

12:12am and 12:21am the 2nd trader dumped 2 bad coins, Grok Bot dumped both too, -$0.48 and -$1.44

1:23am that first coin reached $562,736, Grok Bot cashed the 10x, +$719.68

$149 → $866.76 → $1,612.48 → $1,947.37

7 copies, 5 green, 2 red

I rode 30 seconds behind them the whole night

Be real. At 1:23am who takes profit at 10x, you or the bot

[Joying.eth @Joyingeth]

https://t.co/F3omRjvSpl

[Joying.eth @Joyingeth]

@0xNeuroBlue best 10 repo

[Ridark @ridark_eth]

One agent scoring its own find is a guaranteed yes.
Five that never see each other's work built this $100 desk.

I wired one prompt on Solana. Five agents, each doing one job. A coin moves left to right and any agent kills it.

1. Finder scans every new pair. Under $20k liquidity or past 6 hours: dead.
2. Auditor reads contract and holders. Top 10 above 35% or dev above 5%: dead.
3. Judge scores 0 to 100. Under 80: dead.
4. Stager sizes at 4% of the book. Sets the -25% stop before entry.
5. Closer sells half at double, quarter at triple, trails 30% under the high.

Seven rules sit above every score. 6 open positions, 3 entries an hour, -10% day kills the desk until midnight.

The Finder never knows what the Judge scored. The Auditor reads contracts, not scores. Something goes wrong and I get a name, not a story.

Rule 7 saved the desk. The creator pushed tokens toward the exchange and the Closer cleared at -6% in one block. The coin lost 91%.

[Joying.eth @Joyingeth]

Can't stop f*cking laughing at this scroll.

One pinned canvas, 12 scroll steps, and a full studio portfolio unrolls on a single viewport. 3D cards tilting in depth, glitch letters smearing across the screen, white flushes ripping between sections. Not a single page reload.

Break the scroll open:

1. Pinned canvas. Viewport locks, scroll drives every move.
2. 3D tile carousel. Cards tilt in perspective, swap by scroll.
3. Glitch type layer. Letters smear across the frame.
4. Hard transitions. White flush, then cloud wipe. No fade.
5. Four sections. Fortnite, Unreal Engine, WEAR GO LAND.

The studio behind the scroll is Alche, a Tokyo shop building Fortnite islands in UEFN and Unreal Engine worlds.

GPT-6 Astra builds animated one-shot scroll sites from a public prompt set. The set pins the canvas and warns against scroll hijacking.

Not coding pages. Scrolling them into the viewport.

The card deck tilts and the next work loads in perspective

[Mojunel.eth @mojunel]

YOUR KEYWORD TOOL IS THE PROBLEM

72 keywords entered the scan. 39 survived.

1. Jev pulls the top 10 Google results per keyword. 720 pages land in the queue.
2. Claude fires 8,640 typed judgments across all 720. 216 per second. 204 ms median per result.
3. Forum thread. Dead. Thin page. Dead. Outdated. Dead. Weak domain. Dead. Off-intent. Dead.
4. Keywords with fewer than 3 beatable results in the top 10 get cut. No appeal.
5. 33 keywords drop. 39 survive with 3+ soft slots each.

40.0 seconds. $0.0363 total.

54% of the 72 came back with a beatable spot.
44% of all 720 top-10 results sat on tier-3 domains nobody defended.

https://t.co/qHxPH0MGxR tutorial ranks first. Jev is early access, waitlist only.

Bookmark this

[Joying.eth @Joyingeth]

Can't stop f*cking laughing at this scroll.

One pinned canvas, 12 scroll steps, and a full studio portfolio unrolls on a single viewport. 3D cards tilting in depth, glitch letters smearing across the screen, white flushes ripping between sections. Not a single page reload.

Break the scroll open:

1. Pinned canvas. Viewport locks, scroll drives every move.
2. 3D tile carousel. Cards tilt in perspective, swap by scroll.
3. Glitch type layer. Letters smear across the frame.
4. Hard transitions. White flush, then cloud wipe. No fade.
5. Four sections. Fortnite, Unreal Engine, WEAR GO LAND.

The studio behind the scroll is Alche, a Tokyo shop building Fortnite islands in UEFN and Unreal Engine worlds.

GPT-6 Astra builds animated one-shot scroll sites from a public prompt set. The set pins the canvas and warns against scroll hijacking.

Not coding pages. Scrolling them into the viewport.

The card deck tilts and the next work loads in perspective

[Roan @RohOnChain]

GPT-6 Astra is the most dangerous AI model right now.

It gives you AGI-adjacent reasoning.

That can discover new profitable trading strategies for you 24/7.

If you set it up correctly, you gain a personal hedge fund. https://t.co/Sam7AYOut3

[6AWORON0k @6AW0RON0k]

$44.82 in, $11,643.21 out by morning on a six-agent desk I never wrote a line of code for.

I loaded the Grok Bot terminal last night, dropped $44.82 into it and stopped asking the AI what to trade.

Woke up to $11,643.21. Six agents logged every single fill.

For the past year I kept going back to agents for picks, rereading the same chat threads over and over.

Last weekend I wired up a full trading system on Grok Bot. No hand-written code, ready to run on deploy.

Six agents on one control panel, each with a clear job:
1. Scanner hunts fresh pairs, flags volume spikes.
2. Rug check filters high-risk coins before any order fires.
3. Sizer scales position size with the live balance, never a fixed ticket.
4. Executor opens and closes every position, tracks slippage in real time.
5. Logger writes the reasoning behind each trade into the log.
6. Equity curve refreshes the second a position closes.

Funded the account with $44.82 before bed. Laptop stayed on the dashboard all night.

Opened it in the morning. $11,643.21.

Every order had its reasoning attached, all saved in the log.

That part surprised me more than the balance did.

I stopped agonising over whether to enter. Now I just read the fill log.

Starting balance: $44.82
Current balance: $11,643.21

The whole terminal, the agent stack and the dashboard were built inside Grok Bot.
One deploy and the system runs on its own.