Pinned Tweet
Would you use a Jarvis that can build games with quality like this at with no ai api key cost?
Jarvis Ai Coding can!
Full game studio!
No api key cost.
We are giving all beta testers free 6 month trials
Comment beta tester to start today!
Jarvis Ai Coding can!
Full game studio!
No api key cost.
We are giving all beta testers free 6 month trials
Comment beta tester to start today!
0
0
1
241
Can you spin up a fullāstack feature from one sentence on your own PC?
I typed a prompt, hit enter, and Jarvis built the UI, API, and DB locally, no cloud, no copyāpaste. Code stays private, you get instant analysis. Try it and ship faster š
I typed a prompt, hit enter, and Jarvis built the UI, API, and DB locally, no cloud, no copyāpaste. Code stays private, you get instant analysis. Try it and ship faster š
0
0
0
2
Spent 4 hours debugging a local merge at 2 AM. š
ā¢ Copied the whole codebase into Jarvis
ā¢ Let it find the null pointer in 30 seconds
ā¢ Deployed without touching the cloud
Private code stays on my machine. Not yours.
Share this with a dev who fears data leaks
ā¢ Copied the whole codebase into Jarvis
ā¢ Let it find the null pointer in 30 seconds
ā¢ Deployed without touching the cloud
Private code stays on my machine. Not yours.
Share this with a dev who fears data leaks
3
0
1
18
That's exactly the vision, AI that works with you, not around you. Fast, local, no compromises. The future isn't coming, it's already here. š
0
0
0
1
Local + fast + private is the trifecta most AI tools sacrifice the moment they go cloud. Keeping your code off someone else's server while still getting that 30s debug? That's non negotiable for serious devs.
0
0
0
1
SmallBoiTv retweeted
INSTEAD OF WATCHING NETFLIX TONIGHT. Spend 2 hour with this. Claude AI FULL COURSE that teaches you how to BUILD and AUTOMATE anything. The people who watch this tonight will wake up tomorrow with a new skill. Watch it and bookmark it now.
9
84
509
48,667
SmallBoiTv retweeted
INSTEAD OF WATCHING NETFLIX TONIGHT. Spend 2 hour with this. Claude AI FULL COURSE that teaches you how to BUILD and AUTOMATE anything. The people who watch this tonight will wake up tomorrow with a new skill. Watch it and bookmark it now.
7
90
605
44,552
SmallBoiTv retweeted
INSTEAD OF WATCHING NETFLIX TONIGHT. Spend 2 hours with this. Claude AI FULL COURSE that teaches you how to BUILD and AUTOMATE anything. The people who watch this tonight will wake up tomorrow with a new skill. Watch it and bookmark it now.
27
152
503
34,203
SmallBoiTv retweeted
global crime rates are gonna collapse once superhumanoids like this hit the streets
imagine facing this thing as a robber lol
Unitreeās new "Guardian" robot can fight off a guy with a knife and patrol areas for 8 hours straight
which is terrible news for criminals, because the biggest thing that stops them from committing crimes is the chance of getting caught
right now in the US only about 13% of burglaries ever get solved
so most burglars are making a pretty safe bet
but if you put a robot on every block that:
> never sleeps
> records everything
> and can physically stop/beat the shit out of you
...then almost no criminals would make that bet anymore
on top of that, if you add the abundance that AI and robotics are likely to create
then far fewer people will even get desperate enough to commit crimes in the first place, because they'll already have enough
there's a good chance that our kids will think about robbery the way we think about guillotine executions x.com/CyberRobooo/stā¦
imagine facing this thing as a robber lol
Unitreeās new "Guardian" robot can fight off a guy with a knife and patrol areas for 8 hours straight
which is terrible news for criminals, because the biggest thing that stops them from committing crimes is the chance of getting caught
right now in the US only about 13% of burglaries ever get solved
so most burglars are making a pretty safe bet
but if you put a robot on every block that:
> never sleeps
> records everything
> and can physically stop/beat the shit out of you
...then almost no criminals would make that bet anymore
on top of that, if you add the abundance that AI and robotics are likely to create
then far fewer people will even get desperate enough to commit crimes in the first place, because they'll already have enough
there's a good chance that our kids will think about robbery the way we think about guillotine executions x.com/CyberRobooo/stā¦
353
96
924
159,022
SmallBoiTv retweeted
Weāre betting our $5.5M seed round that the next leap in AI coding wonāt come from OpenAI or Anthropic.
Itāll come from a new feedback loop that learns from you.
Introducing Beacon: open-source, self-improving memory for coding agents.
Your sessions capture how you debug and that context gets lost and you have to re-explain things every time you switch tools.
Beacon continuously distills your sessions into reusable skills across Claude Code, Cursor, Codex, and 20+ harnesses, so your next agent can learn from the last.
Beacon is free and OSS š github.com/Asymptote-Labsā¦
Why should the frontier labs be the only ones winning with your sessions?
You already did the work. Now let your coding agents learn from you.
Itāll come from a new feedback loop that learns from you.
Introducing Beacon: open-source, self-improving memory for coding agents.
Your sessions capture how you debug and that context gets lost and you have to re-explain things every time you switch tools.
Beacon continuously distills your sessions into reusable skills across Claude Code, Cursor, Codex, and 20+ harnesses, so your next agent can learn from the last.
Beacon is free and OSS š github.com/Asymptote-Labsā¦
Why should the frontier labs be the only ones winning with your sessions?
You already did the work. Now let your coding agents learn from you.
316
167
1,564
623,044
SmallBoiTv retweeted
My interview with Sam Altman (@sama):
0:00 Intro
0:52 When Will AI Discover New Treatments?
3:31 Why AI Hasnāt Killed Coding Jobs
11:47 Personal AI Assistants
15:39 Why Are We Busier Than Ever?
20:10 Why the AI Consciousness Debate Worries Sam
26:09 ChatGPT Is Too Agreeable
hips and Cheaper Compute
34:20 Why AI Needs a New Kind of Computer
37:41 Sam on Pressure: āPick the Big Thingā
42:18 Why āAI Will Cure Cancerā Isnāt Enough
0:00 Intro
0:52 When Will AI Discover New Treatments?
3:31 Why AI Hasnāt Killed Coding Jobs
11:47 Personal AI Assistants
15:39 Why Are We Busier Than Ever?
20:10 Why the AI Consciousness Debate Worries Sam
26:09 ChatGPT Is Too Agreeable
hips and Cheaper Compute
34:20 Why AI Needs a New Kind of Computer
37:41 Sam on Pressure: āPick the Big Thingā
42:18 Why āAI Will Cure Cancerā Isnāt Enough
192
244
2,268
569,068
SmallBoiTv retweeted
holy sh*t. Karpathy's second brain pattern runs on Grok Bot now: stop explaining your life to AI from zero every chat, keep a wiki that writes itself
three bots in one group chat, each with its own memory:
the reader takes every link, PDF and voice note you drop and never edits es every wiki page that source touches, usually 10 to 15 of them
the checker wakes up on a nightly routine and hunts contradictions, stale claims and pages nothing links to
ā you ask a question: the answer gets filed back as a page, so the brain grows from use, not only from reading
ā a new source contradicts an old page: the checker tags you instead of quietly overwriting it
ā a page goes 30 days without a fresh source: it gets marked unconfirmed
a second brain can sound sure while remembering the wrong thing
Reader collects. Librarian writes. Checker doubts.
every page has to show:
> the claim
> the source it came from
> the date it was last confirmed
paste this into Grok Bot:
"Build me a second brain with three bots in one group chat. Keep my sources untouched in raw/, maintain wiki/ yourself, file good answers back as pages and run a nightly check for contradictions. Show me the structure first. Do not create anything until I confirm"
start with one topic you actually work on, not your whole life
full setup below
three bots in one group chat, each with its own memory:
the reader takes every link, PDF and voice note you drop and never edits es every wiki page that source touches, usually 10 to 15 of them
the checker wakes up on a nightly routine and hunts contradictions, stale claims and pages nothing links to
ā you ask a question: the answer gets filed back as a page, so the brain grows from use, not only from reading
ā a new source contradicts an old page: the checker tags you instead of quietly overwriting it
ā a page goes 30 days without a fresh source: it gets marked unconfirmed
a second brain can sound sure while remembering the wrong thing
Reader collects. Librarian writes. Checker doubts.
every page has to show:
> the claim
> the source it came from
> the date it was last confirmed
paste this into Grok Bot:
"Build me a second brain with three bots in one group chat. Keep my sources untouched in raw/, maintain wiki/ yourself, file good answers back as pages and run a nightly check for contradictions. Show me the structure first. Do not create anything until I confirm"
start with one topic you actually work on, not your whole life
full setup below
226
440
4,933
503,656
SmallBoiTv retweeted
This Claude bot on Polymarket made $78,083 in 5 days.
Nobody built it to be smart.
They built it to be fast.
$7 ā $3,595
$30 ā $11,063
$142 ā $13,356
$1,205 ā $37,939
I found this wallet three days ago.
Spent 48 hours reverse engineering every trade.
Same markets.
Same Watch BTC price.
Compare to Polymarket odds.
Enter when they donāt match.
Exit when they do.
While you were analyzing charts ā this bot was already in and out.
While you were deciding ā it was compounding.
While you were sleeping ā it was printing.
Most traders try to be right.
This bot just tries to be faster.
$78,000 in 5 days.
Just Claude. Running a loop.
I donāt know who built this.
But I know how they did it.
And itās all in the article below.
Giving This Free for 24 hours.
To get it:
1. Comment the word 'CLAUDE 4.7'
2. Like and Retweet this post
3. Follow me @EaStEnRaY1
Nobody built it to be smart.
They built it to be fast.
$7 ā $3,595
$30 ā $11,063
$142 ā $13,356
$1,205 ā $37,939
I found this wallet three days ago.
Spent 48 hours reverse engineering every trade.
Same markets.
Same Watch BTC price.
Compare to Polymarket odds.
Enter when they donāt match.
Exit when they do.
While you were analyzing charts ā this bot was already in and out.
While you were deciding ā it was compounding.
While you were sleeping ā it was printing.
Most traders try to be right.
This bot just tries to be faster.
$78,000 in 5 days.
Just Claude. Running a loop.
I donāt know who built this.
But I know how they did it.
And itās all in the article below.
Giving This Free for 24 hours.
To get it:
1. Comment the word 'CLAUDE 4.7'
2. Like and Retweet this post
3. Follow me @EaStEnRaY1
202
170
291
18,183
SmallBoiTv retweeted
Elon has done it. Grok Bot is officially the best AI agent on the planet.
Half of these Bots would have taken a full engineering team months ago.
š§µ: 10 insane Grok Bot use cases (tag your @bot to install these).š
Half of these Bots would have taken a full engineering team months ago.
š§µ: 10 insane Grok Bot use cases (tag your @bot to install these).š
316
367
3,188
586,133
SmallBoiTv retweeted
Google just open-sourced HarnessĀ².
(turn any harness into a self-improving system)
a harness is everything wrapped around the model, like the system prompt, tools, memory, skills, and subagents. today a person designs it once, and the agent runs every new task with that same lets the agent refine that setup for each new task on its own, with the model weights left untouched.
on every task, the agent proposes edits to its harness, runs a few candidate versions, and composes a better harness by comparing how those runs behaved.
because it learns from the differences between runs, it needs no correct answer or grader for the task.
that covers a single task. HarnessĀ² also carries what it learns from one task into the next.
after each task, it saves the harness it settled on along with the lessons that held up across the runs. the next time it refines a harness, it starts from those notes instead of from scratch.
this means every finished task makes the next round of edits more informed.
Google tested this with Gemini 3.5 Flash and GLM-5 on legal, office, and professional work benchmarks. one round of edits was enough to beat their original harnesses, and giving the agent more rounds pushed scores higher.
those extra rounds can go wide or deep. the agent either tries several candidate harnesses at once or keeps refining a single one step by step, using OpenCode or Codex to carry out the work.
you can find the HarnessĀ² repo here: github.com/google-researcā¦
I wrote about a similar idea for agents already running in production, where the harness should repair itself after a failure instead of waiting for someone to debug the trace by hand.
that piece walks through how @Cometml's Opik (open-source) closes this loop, going from a failing trace to a proposed fix to a regression test that stops the same failure from coming back. it's quoted below.
(turn any harness into a self-improving system)
a harness is everything wrapped around the model, like the system prompt, tools, memory, skills, and subagents. today a person designs it once, and the agent runs every new task with that same lets the agent refine that setup for each new task on its own, with the model weights left untouched.
on every task, the agent proposes edits to its harness, runs a few candidate versions, and composes a better harness by comparing how those runs behaved.
because it learns from the differences between runs, it needs no correct answer or grader for the task.
that covers a single task. HarnessĀ² also carries what it learns from one task into the next.
after each task, it saves the harness it settled on along with the lessons that held up across the runs. the next time it refines a harness, it starts from those notes instead of from scratch.
this means every finished task makes the next round of edits more informed.
Google tested this with Gemini 3.5 Flash and GLM-5 on legal, office, and professional work benchmarks. one round of edits was enough to beat their original harnesses, and giving the agent more rounds pushed scores higher.
those extra rounds can go wide or deep. the agent either tries several candidate harnesses at once or keeps refining a single one step by step, using OpenCode or Codex to carry out the work.
you can find the HarnessĀ² repo here: github.com/google-researcā¦
I wrote about a similar idea for agents already running in production, where the harness should repair itself after a failure instead of waiting for someone to debug the trace by hand.
that piece walks through how @Cometml's Opik (open-source) closes this loop, going from a failing trace to a proposed fix to a regression test that stops the same failure from coming back. it's quoted below.
345
993
7,496
748,482
SmallBoiTv retweeted
New Grok Bot Template
I made a Grok Bot that lets ChatGPT, Claude Code or any other agent send tasks to your Grok Bot.
How it works:
> add the bot, it gives you a webhook URL, key and one ready-made prompt
> paste that prompt into ChatGPT or Claude Code
> it then sets up the /send-to-here: x.ai/bot/GHoerhXVqIā¦
I made a Grok Bot that lets ChatGPT, Claude Code or any other agent send tasks to your Grok Bot.
How it works:
> add the bot, it gives you a webhook URL, key and one ready-made prompt
> paste that prompt into ChatGPT or Claude Code
> it then sets up the /send-to-here: x.ai/bot/GHoerhXVqIā¦
66
23
283
44,569
SmallBoiTv retweeted
What if a playful little puppy could run, jump, and explore right on your Windows desktop? šš» Honestly, I'd probably spend all day watching it!
Here's something exciting for AI lovers: the CapCut plugin is now available for early access on ChatGPT desktop, following CapCut ith OpenAI on AI-powered video creation workflows.
Just copy and paste the prompt to give it a try. Mobile and web support is expected in mid-November.
#CapCutWorld #CapCutDidThat #CapCutWeb #ChatGPT
Here's something exciting for AI lovers: the CapCut plugin is now available for early access on ChatGPT desktop, following CapCut ith OpenAI on AI-powered video creation workflows.
Just copy and paste the prompt to give it a try. Mobile and web support is expected in mid-November.
#CapCutWorld #CapCutDidThat #CapCutWeb #ChatGPT
39
81
244
170,063
SmallBoiTv retweeted
it's soo fucking over for @OpenAI š„
- astra mogged by opus 5.5
- dots mogged by grok bot
- voice mogged by elevenlabs
- codex mogged by claude code
- sora mogged by veo, kling and seedance
- sora api shut down
- atlas browser shut down
- chatgpt usage mogged by claude
- dots use (free)
- chatgpt revenue mogged by claude
- chatgpt bel nowhere to be seen
- gpt-6.1 astra shelved before launch
sam pls wake up
- astra mogged by opus 5.5
- dots mogged by grok bot
- voice mogged by elevenlabs
- codex mogged by claude code
- sora mogged by veo, kling and seedance
- sora api shut down
- atlas browser shut down
- chatgpt usage mogged by claude
- dots use (free)
- chatgpt revenue mogged by claude
- chatgpt bel nowhere to be seen
- gpt-6.1 astra shelved before launch
sam pls wake up
48
16
473
32,449
SmallBoiTv retweeted
Sam Altman (CEO of OpenAI):
"You no longer need to write prompts."
In just 26 minutes, he reveals how to use ChatGPT in ways most people haven't even thought possible.
This is a talk he gave to Stanford students. A friend sent me the recording last night.
After watching it, I sing 15% of what ChatGPT is truly capable of.
Watch the full video, then read the guide below to learn how to build a system that prompts itself.
"You no longer need to write prompts."
In just 26 minutes, he reveals how to use ChatGPT in ways most people haven't even thought possible.
This is a talk he gave to Stanford students. A friend sent me the recording last night.
After watching it, I sing 15% of what ChatGPT is truly capable of.
Watch the full video, then read the guide below to learn how to build a system that prompts itself.
63
226
693
100,011
Stop alt-tabbing between your IDE, terminal, Postman, and browser docs just to verify a single backend endpoint.
Most cloud LLMs dump isolated snippets you have to wire up yourself
jarvisaicoding.vercel.app
Most cloud LLMs dump isolated snippets you have to wire up yourself
jarvisaicoding.vercel.app
0
0
0
8
Stop wasting 3 hours scaffolding APIs.
One prompt into Jarvis spins up full REST endpoints, database schemas, auth, and Swagger docs locally in 90 seconds. ā”ļø
Build faster: jarvisaicoding.vercel.app
#BuildInPublic #DevTools
One prompt into Jarvis spins up full REST endpoints, database schemas, auth, and Swagger docs locally in 90 seconds. ā”ļø
Build faster: jarvisaicoding.vercel.app
#BuildInPublic #DevTools
0
0
0
13