🔥 https://xitter.cf/t.co/vYTWZL0StE - Im creating the vibe coding environment of the future ( macOS app ) 📺 YT -https://xitter.cf/t.co/lrpw6a0FQQ
Joined July 2017
- Tweets13.8K
- Following1.1K
- Followers24.9K
- Likes16.9K
Pinned Tweet
if vibe coding was actually a VIBE :
yes you can run remote canvasses straight on your vps and ship to production like a psychopath.
yes its built in swift and uses very little memory and is insanely performant.
yes it has agentic voice control with gpt realtime 2.
yes their is bidirectional agent / canvas control via mcp and cli.
yes their is built in shared memory system without bloat.
made with love by a dad in his basement who is tired of vibe coding being slow, boring, and constrained.
CNVS coming soon.
Eveyone has been waiting for anthropic to enter the personal agent game.
opus 5.5 mogs everything out rn.
the polish / soul / vibe in claude products feels unmatched in the ai world.
apple level.
expectations are going to be insanly high for claude duet and I cannot wait to get my hands on it.
The future just got brighter.
Hermes and Openclaw will always have the advantage of truly owning your agent, duet will have to provide a word class experience to overcome this.
Would you rather own a hermes / claw or receive a done for you bot / dot / muse / duet?
I had my first long form youtube video get 50k views edited entirely by opus 5.5
from my livestream
The problem is the agent receives a direct copy of the stream, meaning no individual layers ( face cam, desktop, audio tracks )
this limits the agents ability to make a banger video.
I just overhauled my setup and now the agent receives clean layers in 4k for screen recording, full face cam, and independent audio tracks.
i cant wait to see what ai is able to do with this.
we are soo early.
im trying to build capacity now.
( btw yesterdays 4 hour stream was 570gb so i still have some things to figure out ... )
inspired by @rileybrown and the insane work he has done with short form ai editing.
My opus 5.5 agents are nocturnal.
during the day i plan and strategize wildly ambitious goals that can be measured.
before bed i setup the god prompts.
worker agents and review agents go round for round and absolutely destroy my claude 20x accounts.
i wake up somedays to masterpieces
and sometimes dissapointed. But with opus 5.5 most days i wake up blown away. Like this nuketown map clone literally broke my brain.
runs in the browser, you play it at the link below.
The biggest paradox in ai right now :
“What is easy to do, is also easy not to do.”
Jim Rohn said this in the 60s.
Opus 5.5 has made everything easy, almost anything digital is a prompt away.
People still won’t take action.
If you want to win you just have to keep showing up, keep prompting, stay consistent.
This is UNBELIEVABLE 🤯
1. The shoutout from Riley is insane
But more importantly
2. These shorts he is making with a 3d BLENDER environment are ridiculous.
natural story telling and communication skill COMBINED with the ai editing, animations, and scene creation is unstoppable.
I couldn’t be anymore fired up to experiment with my own version of this.
Low effort Slop content will FLOP.
You have to push the limits and use ai to produce content that make people go “wtf”.
This video was edited with 1 prompt with Opus 5.5.
I uploaded a single video of myself (with many mistakes). Opus went on to:
1. Analyze the transcript
2. Search Twitter for the relevant B-roll (Claude Code literally calls GrokBot for this step)
3. Scrape the Twitter video assets
4. Cut the video based on mistakes and word gaps
5. Build out the Blender scene from a template
6. Build the UI and components of each shot (my Blender scene has 4 shots: a computer, phone, table, and board)
7. It took screenshots of every section, analyzed and fixed them
8. Synced every shot, zoom, and animation to the exact words I say
9. Animated a robot arm placing real logos on the table right when I mention them
10. Drew the whiteboard ending with logos and graphics instead of text
11. Pulled a cutaway from the creator's own video, with his voice
12. Rendered all 3,487 frames on RunPod cloud GPUs in under 20 minutes, for less than $1
13. Put me in the bottom third only when I'm talking, so the B-roll stays full screen
14. Added word-by-word captions, sound effects, and music
15. Uploaded it to my Google Drive, ready to post
Will be making a full youtube video this week explaining it. I want to learn a little more about how cloud rendering works before making video, as well as figuring out what the best provider is. (If you are in cloud rendering space, hit me up)
My goal of the youtube video will not be to make it a copyable skill, rather teach people the foundations so they can be creative and make their own mini-world by which they can make videos from, while understanding what's happening under the hood.
Shout out @_chuckyc @_MaxBlade
yesterday i launched my hivecoded battle royale for a live playtest event.
Kong Royale.
we did 6 lobbies of 50 players and EVERY GAME was 100% full.
Everything just worked.
it took 3 weeks, 2 claude max accounts, and 100s of prompts from my livestream chat to build.
Even when players disconnected, opus 5.5 built a reconnect feature to get them right back in the lobby.
p2p in game voice chat was incredible.
I had 100fps all day, and server latency was around 20ms.
people played from all over the world.
i flew helicopters, drove a tuk tuk, and had intense battles trying to rotate around the map.
you CAN do anything now.
Pacing around my house yelling “no fucking way.”
Elon does it better.
Hearing “No self driving” felt good on the soul. Occasionally we need to keep things manual, exciting, and fun.
Just because you can automate doesn’t mean you should.
The future is looking incredible rn.
I have spent 3 weeks and two claude max accounts vibe coding a 50-person battle royale.
Kong Royale.
We invented hive coding, chats messages prompt 20 agents working overnight to fix and improve the game.
and today we finally launch 🦍🍌
live playtest TODAY 1 PM ET. drop in with us 👇
youtube.com/watch?v=b5n1K_IA…
Can gemini 4 argon 3d print a working fidget spinner?
lets find out
this is one of the hardest benchmarks I have given to these models. Opus 5.5 has failed, Astra has failed, and yes even grok 4.7 lol
Will gemini finally let me print a toy for my kids?!
i am using a bamboo labs p1s
The only constraint is it needs to print in under one hour.
okay here is Argon vs Wonder vs Opus
"build me a lawn mowing simulator i can smoke a cigar and drink a beer in... "
Argon performs a decent amount better than wonder here, but still is not in the same galaxy as opus.
An engineer at Gemini put out a beautiful video explaining how important these 3d demos are to showcasing and bechmarking these general intelligence models.
I will link it below (:
"3d builds are a great way to test a models capabilities particularly in areas like coding.."
xitter.cf/Google/status/21100412…
Argon brings major improvements in spatial reasoning, mathematical precision, and visual coding.
Using multi-agent loops in @Antigravity, it autonomously built this entire pagoda complex, sculpting the architecture, foliage, and lighting from scratch in Blender.
Is Muse Wonder full better than opus 5.5?
I gave them the same prompt
goal is to build me a triple aaa level lawn mower simulator i can drink beer and smoke a cigar.
Api equivalent cost :
muse - $4.11
opus - $99.43
It is very clear that muse is not anywhere near close to frontier in 3d tasks like video games, or even printable cad designs.
Even scarier is it burned my $50 plan spending $4.11 in api equivalent tokens.
wut?!
i had to upgrade to $200 plan to finish this run.
I built both games AND the comparison video in my ADE -> cnvs.dev
This made ALOT of people angry so let me explain myself :
1. Are you worried about prompt injection?
No, opus 5.5 is ridiculously good and my agents run in a remote environment on a worktree.
2. Does this cost $15k a month?
No opus 5.5 with subagents is insanely cost efficient I have two 20x max accounts, $400 a month. I make this back in product sales in two days.
3. This is just an agent swarm?
No this is the opposite. A human swarm ( 100s of people ) prompting a few agents. This is the future of open source, don’t bring the code, bring the idea / prompt.
4. How do I play the game?
Battle royale goes live on Thursday ( 100% free ) you must sign up here -> kongroyale.com to get access
5. What ade do you use?
I built my own -> CNVS.dev
Hivecoding is the most powerful use of Ai that i have ever used...
25 opus 5.5 agents + top chat messages = hivecoding
Everyday i live stream and 100s of agentic engineers type fixes and improvements in my chat.
most people in my chat are real devs and much smarter than me.
i have agents run overnight, filtering out the best improvements and suggestions from chat and from my stream transcript. Last night my agents shipped 90 new changes to Kong Royale.
( 50 person battle royale im building )
i wake up to MASSIVE improvements everywhere.
the agents make changes to the stream, my production setup, our discord, etc...
The human is now the bottle neck, but not when you have 100 to 200 humans prompting your agents everyday.
