YouTuber, Educator, Founder Building an agent operating system @chorus_agent Updates on the best agent tools @aisuperapp
in many cases it's worth it.
RT Emily Lambert I used Fable 5.2 to build an air traffic control game where the planes are flowers. lvl 1 starts out very peaceful. It does not stay that way. (Voices and sound design with @ElevenLabs + Epidemic Sound 🌸)
RT Yoni Smolyar Holy
The Fable of Clawd A 2 minute 4K film made entirely in Blender, from one prompt on Fable 5.2. It took about 4.5 hours. Claude built every asset in Blender by writing Python: - 5 characters and props: Clawd, his friend Pip, a toy truck, a 3D-printed tulip and a gift box - 7 full
View quoted postepisode 2 tomorrow (it's 10x better)
This is actually INSANE. Fable 5.2 just created this in one shot, and I did it live on stream. I gave Claude one prompt. It wrote code that builds the entire room in Blender: the desk, the MacBook, the light, even Clawd. The laptop screen is a web page Claude built to look
View quoted postin 2035 there will be 2 software companies left. Anthropic and Bending Spoons.
2025 OpenAI lost focus, Anthropic took the lead 2026 Q1-Q2 Anthropic lost Focus, OpenAI took lead 2026 Q3-Q4 OpenAI lost Focus, Anthropic took lead
1000%
Would you enable yolo mode for dots, even if your dot might do sth unwanted?
View quoted postdude....
Introducing Claude Meetings: add Claude to your team’s video call and it builds docs, whiteboards, and other artifacts with you while you’re still talking. Starting to roll out in beta on Team and Enterprise plans.
View quoted postFor a film I’m making in NYC I need a pool for an underwater shot for 5 hours. In the next few weeks. Can anyone help with this?
if you are ambitious do everything in your power not to have a "job in social media". Build your own thing as fast as you possibly can. Become your own brand, and watch your entire world change.
Wow. This is a big deal. I track social media salaries by industry and sports is one of the lowest paying.
correct.
refuse to be audience captured. always be exposing your audience to things they don’t gaf about. test their loyalty to the follow button. LISTEN TO THE NEW 2HOLLIS ALBUM
View quoted postWho has cracked sound design with Fable 5.2 It's the missing piece
wow it feels so good to be freed from menus. whenever i need to change settings in any app i just ramble what i want to codex and it uses computer use or browser use to set it up. Yesterday i had it fully set up a discord community with proper settings and security and it just flew through it. if you haven't tried codex computer use esp with terra, i highly reccomend it. I would guess we are 6 months away from AI being faster than us at controlling a computer.
RT Riley Brown Fable 5.2 is the best model ever released.. and it's not close. Especially at any visual task. Here are the 6 reasons why: (This entire video was edited by Fable 5.2 in 40 min) If you want to use the same intro style as me i'm putting the github-for-video link down below. Just copy and paste it into claude and tell it to edit a video just like it. 00:00 Intro 00:41 1. Blender & 3D 06:24 2. Video Editing 09:55 3. Video Games 13:13 4. Apps Are Solved 15:08 5. Research Decks 16:26 6. Vibe Coding Hardware
I've been waiting to make my agentic video editing series for a while... It's time. Oh my god fable 5.2 is the breakthrough. It's funny my first instinct is i need to hire 5 editors to use these tools to scale content. That's how much leverage these tools will give creative teams. I will make many videos over the coming weeks covering where specifically the leverage will come from.
RT Robert Scoble Very nerdy, but I love it. Muhammad updated his old Jeep with AI so it would pass smog.
Claude reverse engineered my 32-year-old Jeep. Then it started controlling it. Edited with Fable 5.2
View quoted postThis is the coolest video i've seen in a very long time.
Claude reverse engineered my 32-year-old Jeep. Then it started controlling it. Edited with Fable 5.2
View quoted postearly feedback @AnthropicAI please allow to paste images with no background without the little tape thing would love to be able to past logos, icons etc without having white bg
Starting to roll out today: Claude Whiteboards, Claude Sheets and Claude Data. Think a plan through on a board, build a tracker in a sheet, and tell a story with your data.
View quoted postthe pelican on a bike benchmark needs an update. the picture: one year ago the video: today https://x.com/techfren/status/2119924958134702276/video/1
GTA 7 will be a claude artifact
Fable 5.2 Yesterday. Gemini 4 Carbon Today.
Gemini 4 Carbon, now available in the Gemini API and to Ultra subscribers! Just a few weeks after Argon was launched, this model makes significant progress on lots of the feedback from Argon and pushes the frontier in cyber, knowledge work, and code! It's a great model : )
View quoted postWhat are the best theories on how Anthropic made their models so damn good at visual understanding and taste? Fable 5.2 (and even Opus 5.5) isnt just good at front end design or Blender… its great at making anything visual, in ANY tech stack. It feels like AGVI (Artificial General Visual Intelligence) How, How, How?
hahahha omg
I wanted to see how far i could push Fable 5.2 I gave it two claude max accounts and access to unreal engine 5.9 beta mcp it ripped for 4 hours. it made the most horrifying terror game i have ever played.
View quoted posthuge huge huge
Dots can now control many machines in addition to its own little cloud computer that we provide by default.
View quoted posttomorrow will be fun too!
Here's a video edited by Fable 5.2 about a video I edited with Fable 5.2.
This is actually INSANE. Fable 5.2 just created this in one shot, and I did it live on stream. I gave Claude one prompt. It wrote code that builds the entire room in Blender: the desk, the MacBook, the light, even Clawd. The laptop screen is a web page Claude built to look
View quoted postRT Roberto Nickson Uploaded a few pictures of my studio and ran a similar prompt as Riley. One-shotted by Fable 5.2 in around ~2 hours. Wow.
This is actually INSANE. Fable 5.2 just created this in one shot, and I did it live on stream. I gave Claude one prompt. It wrote code that builds the entire room in Blender: the desk, the MacBook, the light, even Clawd. The laptop screen is a web page Claude built to look
View quoted posthey someone at @runpod can you hook me up with credits, i'll bring you thousands of users. thanks.
i've never once found these charts helpful. i'm convinced it's all pretend. trust your own tests, your own eyes.
There is a skill circulating called REA-Video "Reverse Engineer Anything - Video" Essentially claude downloads a video from youtube and analyzes every frame and the relationship between them to "extract" the video editing style. And i'm afraid with Fable, this is VERY good now. I've tested it on Kurzagets video... i did it on stream, i dont think i'm gonna post it for the negative backlash from the editor community, but i just think people should be aware of where this is going. @elder_plinius tweeted the following the other day: my spidey senses are telling me the labs are, as we speak, using unguardrailed versions of their unreleased models to systematically reverse-engineer ALL of the internet’s software, apps, and games, then feeding the reconstructed source code straight into their training pipelines while simultaneously building guardrails to ladder-pull the rest of us I have a feeling the path to creating a model that's VERY good at editing and creating videos is something like this, but with Videos. Reverse engineer every good video to code files then use that data to train models. again i'm not saying any one is or anyone should do this. just... a thought.
Dear Startup Founders of San Francisco. Request for a startup: a rendering platform ONLY for video editing. Figure out the fastest way to render videos and let claude and codex plug into it, and it renders any type of video and returns a download link with the video. AS FAST AS POSSIBLE. I have a theory that in one year most video editing software will be bloat. And that the best editor will be claude code and codex editing the files headless. These videos will need to be rendered somehow. The better the models get the more aspirational people will get with their videos. Creating an 20 minute video that renders in 8k on blender at the highest quality should be no problem. There are alot of cloud providers that technically offer this but they are sloppy for this use case, and drain tokens unnecessarily.
this model is an absolute unit. lightyears ahead anything i've ever seen in terms of visual understanding. I'm astounded. this thing has better taste than i ever will.
Amazing feature. well done @X
Clipping is here 🎬 Grab your favorite moment from any video, add your take, and post it without leaving X. Creators: every clip credits your original video and links back to it, so new viewers can find your work and keep watching. Clippers: you now have an easy, native way
View quoted postfuck my computer is melting omg what is happening.
RT Farzad 🇺🇸 🇮🇷 BRO
This is actually INSANE. Fable 5.2 just created this in one shot, and I did it live on stream. I gave Claude one prompt. It wrote code that builds the entire room in Blender: the desk, the MacBook, the light, even Clawd. The laptop screen is a web page Claude built to look
View quoted postRT Zero Foundry Saw this one live. HOLY.
This is actually INSANE. Fable 5.2 just created this in one shot, and I did it live on stream. I gave Claude one prompt. It wrote code that builds the entire room in Blender: the desk, the MacBook, the light, even Clawd. The laptop screen is a web page Claude built to look
View quoted postFable 5.2 is here https://x.com/i/broadcasts/2vylypjRlNNAAJ
badaboom badabing
RT Anthropic Claude Fable 5.2 and Claude Mythos 5.2 are available now:
Introducing Claude Fable 5.2 and Claude Mythos 5.2, designed to tackle the most daunting technical challenges. They’re capable collaborators on ambitious projects in many fields, and show improvements in areas like coding, reasoning, and vision.
View quoted postIt feels like nothing has happened in AI since Opus 5.5. But... that is about to change. The next 2 weeks will be very very fun.
"okay can you please make a vertical version of this that looks like a UGC video"
One shot with Opus 5.5 with my new blender video editing workflow done in less than 10min. "Create a launch video for an app just like screen studio but showing a desktop app and iOS app"
View quoted postClaude Duet can now connect to Claude Code running on my Mac. This is useful for these 5 reasons 00:00 Intro 00:22 1. Virtual + Physical Computer 03:24 2. Edit Videos From Your Phone 06:56 3. Build Apps From Your Phone 08:48 4. Computer Use From Your Phone 10:29 5. Real-Time Voice With Claude Code
Claude Duet can now Control Claude Code that's running on your computer. This will be great for video editing and coding tasks. - You can control claude code running on multiple computers - You can work on existing sessions - You can spin up new sessions - You can do this with real time voice (i hope this get's better, their real time voice is mid)
Claude Duet: now with Claude Code sessions Check what's running, approve changes and send updates - all from your phone. Works with sessions on your computer or in the cloud. /cc @navbryce and @joshleibsly
View quoted postnever been a better time to be a creative. today while in an uber, i had claude duet use it's vm to gather assets for a video idea i had, i then had duet spin up a claude code session running on my mac mini at home to edit the video using blender for a few visuals. To render the video faster claude code rented compute by the minute on runpod and rendered the blender scenes, then it finished the render on the mac mini, uploaded it to my video version control app and sent me a link on duet where i reviewed the edit. I went back and forth with claude duet 5 times in the matter of 30 minutes until the video was done. Pretty insane process and was just chatting with an agent the whole time. this is how i storyboard my video ideas. I make ai mockups of the video before i ever film.
That theo guy bouta raise like a trillion dollars for t3code isn't he?
There are a lot of videos going around X on how to grow on social media. Most of them are very zoomed out and look at the broader trends. If you’re just getting into marketing, I think that’s the wrong place to start. You should be studying how to make one good video. Can you write a script, film a video, and edit it, so that it provides value and goes viral? If not keep trying over and over and over and over until you develop that instinct. The ability to just create valuable content on demand... That is your goal. Think of content like a startup. Just make a video people want. Once you find it, congrats you have CMF (Content Market Fit). Okay do it again and again and again and again and again and again improving every time. (btw this guys video was good. But if you don’t know how to make a good video, none of it matters)
if u want access to what i believe is the best free video editing tool that's releasing next week comment on this post. Works directly inside claude code, codex, and grokbot. again 100% free and open source.
You have an always on persistent agent powered by Opus 5.5 with it's own computer... That can also control Claude Code on your computer. I don't think people realize how powerful that is.
Claude Duet: now with Claude Code sessions Check what's running, approve changes and send updates - all from your phone. Works with sessions on your computer or in the cloud. /cc @navbryce and @joshleibsly
View quoted postRT Jake Gold Me too but right now AI bots are being held back by...the world. Agents just can't do a lot of stuff, for a bunch of reasons. It took many years for the world to become smartphone-native. It will take at least a couple years for the world to become agent-native. Then crazy.
i expect my dot to get VERY good.
Some news: I’m joining OpenAI to lead product management for Core Products! I have a lot of respect for OpenAI for showing the world that scaling works, for bringing AI to over a billion people, and for believing great products are core to the mission. I’m excited to work with
View quoted postRT Julie Chen i made a list of the best tech storytellers on X in 2026 @cosign the bar: – builds agent-native. no commentators – at least one viral post this year (1M+) – can turn a product into a narrative people repeat storytelling is the moat https://cosign.co/lists/0xjuliechen/best-tech-storytellers-on-x
RT djcows "the AI bubble will pop soon" - guy who tried gpt-3.5-turbo once in 2022
n̶e̶w̶ ̶b̶u̶s̶i̶n̶e̶s̶s̶:̶ ̶t̶e̶a̶c̶h̶i̶n̶g̶ ̶p̶o̶d̶c̶a̶s̶t̶e̶r̶s̶ ̶h̶o̶w̶ ̶t̶o̶ ̶k̶e̶y̶ ̶w̶o̶r̶d̶ ̶o̶p̶t̶i̶m̶i̶z̶e̶ ̶f̶o̶r̶ ̶d̶i̶s̶c̶o̶v̶e̶r̶y̶ new prompt for claude: i'm a podcaster, what words do i say to optimize for discovery.
new business: teaching podcasters how to key word optimize for discovery
View quoted postRT John KM (YouTube Strategy) Been seeing some ridiculous results with @rileybrown of late So many bangers 🔥
YouTube firing on all cylinders. My videos from the past 10 days. More guest episodes coming up soon. Going to be digging deeper in the stack, talking about Open Models, Infra, and deeper business workflows and more.
YouTube firing on all cylinders. My videos from the past 10 days. More guest episodes coming up soon. Going to be digging deeper in the stack, talking about Open Models, Infra, and deeper business workflows and more.
This is currently my fav vlogger on YouTube. 19 year old kid just moved to NYC for a gap year. His vlogs are incredibly polished. A good mix of slow vibes with alot of cuts and good visuals and music. Each video feels like a mini documentary. 10th video on his channel and already at 14k subs. I predict he'll be at 250k subs in 2 months.
People seem to think its "silly" that vibe coders are going to be maintaining apps for themselves. My take is that its not silly at all... GPT Sites has built in Auth, DB, Storage, Hosting... everything... GPT Sites will continue to get better and easier to use. Once you build your site just prompt it: "every day please make sure this site is secure and maintained properly, make any changes you need to" Done! You now have a personal software application that is auto-maintained for YOU! This is already better than most SAAS. Why? Because SAAS companies do not maintain software for YOU. SAAS companies maintain software for it's current and future pool of users, and their business model first. You as an individual are not prioritized. And there's also a common misconception that we are a bubble on x. I've seen it with my own two eyes. The most "normie" of "normies" are building and successfully using software that they build themselves. And we are only in like inning 2/9 on this. It will only accelerate.
RT Suhail Now in OpenClaw - it was collaborating with them on this! We have a lot more updates to the model coming...
Particle-1 is available in OpenClaw. @particle_api reports ~45% fewer output tokens than DeepSeek V4.1 Flash with comparable benchmark accuracy. Zero data retention for standard inference. To give it a try, update to OpenClaw v2026.10.10. https://particle.ai/blog/introducing-particle-1
Your Agent can now search through millions of podcast transcripts. Whether you use Claude Code, Codex, Grokbot etc... This is useful for many reasons. The first reason is it allows you to find better podcasts to listen to. But also. this allows your agents to grab a perspective before doing a given task. If you ask it to create a landing page, or create content script or anything else that requires a cultural understanding you can send it to a specific podcast or group of podcasts and this will ground your agent in a perspective. This will improve its tone, understanding of any niche, and the accuracy of the response. I've been using a custom built solution for this for months and it kinda sucked... but now it works great and its fast... this video feels sponsored but it's not.... just genuinely useful.
this API is incredibly useful. i don't think i'm going to rely on any app to recommend podcasts anymore. i'll just have my agent search thousands of episodes, transcripts included, and recommend what i should listen to.
this API is incredibly useful. i don't think i'm going to rely on any app to recommend podcasts anymore. i'll just have my agent search thousands of episodes, transcripts included, and recommend what i should listen to.
Exa now indexes 3,000,000+ podcasts. Your agents can now search across millions of hours of transcripts, updated daily, and track what the world is talking about.
View quoted postThis is cool but a single 90min YouTube video released by Karpathy would be more valuable than this entire site.
Weekend hack: HackerTube. I was annoyed with YouTube's recommendations so I had Claude filter down 43M comments over 20 years of HN to 6,454 videos that keep being recommended. *Significantly* better viewing experience, ~2 hours with Opus 5.5: https://hackertube.org/ - Excellent
Have you used the Muse app in the past 7 days.
My only regret is not moving to New York sooner.
Watching long form youtube videos doesn't feel better on your brain than watching short form reels because good youtubers basically turn a long form video into a series of short form videos. Hook, Rehooks, open loops, payoffs etc. And the main hook needs to be more potent than short form hooks because it needs to set the stage for a longer video. This is not the creators fault either. The algorithm simply optimizes for this.
RT Zack Voell Dudes will dress up for Halloween at 29 years old while actively in Klarna debt
RT Emily Lambert How it started vs. how it’s going. I’ve had so much fun designing our studio. So much of the work is figuring out what belongs in the room and what doesn’t, while making it functional and somewhere we actually want to spend time. Still a work in progress, but I’m proud of it.
8 minutes...
Granola now does voice typing. Unlimited words, for free. Granola already writes your meeting notes. Now it can do the typing for you, in any app.
View quoted postfind a direction then protect your attention.
Within 12 months both meeting notes and voice typing will be completely free.
Granola now does voice typing. Unlimited words, for free. Granola already writes your meeting notes. Now it can do the typing for you, in any app.
View quoted postNot necessary, waste of focus. You can set up a gpt-image-2 skill on Claude in 20 seconds.
As of today October 29th. This is my current AI Stack. Mostly comprised of 2 agent tools, and 1 that allows me to centralize my plugins and skills which makes it easier to switch agent platforms. https://youtu.be/YT0u2eQ-Aj4?is=eRXb-ePvR8EnoGoo
Either the platforms remove the ability/incentive to post more (they won’t) or it will become REQUIRED to post hundreds of videos per week to build your personal brand.
This is cognitive pollution and we are drowning in it already, and people think this is good. MORE CONTENT
View quoted postCan someone at @OpenAI help me text my dot. I don’t see the option. I’m on a team account with multiple seats and I’m wondering if there’s a setting I need to enable.
Can i please have access. my username is @RickRoss (not kidding)
The @suno MCP (beta) is here and available for early access! Many already plan and work on songs on other platforms and have asked for us to build a connector so they can work without switching between tabs. We're looking for early testers to use the MCP and provide feedback --
View quoted postThis is A++++++++++
We partnered with Suhail to create the brand identity for his new company, Particle.
View quoted postIf you have strong intuitions about the future, have high agency, and are creative... you need to put yourself in a position where you can act on those intuitions. Ideally with enough freedom to navigate and follow those intuitions as they change. 1 year in the singularity is worth 10 normal years. Think about the opportunity cost of spending a year in something you don't deeply believe in.
we'll see when we test it. These charts dont matter
I don’t want touch screen. I just want built in 5G.
We brought Touch to the Mac! From the all-new iPad mini to the 14-inch MacBook Pro and iMac... and now MacBook Pro with PROLED XDR Display and Touch, creators have new tools to do their best work yet.
View quoted postDoes anyone see this option in app?
By popular request, you can now text your dot. Give it a task to take off your plate or something to keep an eye on, using text, photos, or even voice notes. It will text you proactively with relevant updates. Safe, secure texting via iMessage or RCS. You’re the only one who
View quoted postyes! finally someone else is talking about it. dot + 24x7 mac is the elite setup. you can literally call your dot and with only voice do ANYTHING on your computer. just by chatting. hyped for the blog post.
openai dot and a 24/7 mac mini have made my life 15% better. couldn’t recommend more. will write a blog post on my setup
View quoted postYeah keep in mind this is all for work. I want to agent max in my professional life but tbh for my personal life I want to read, write and play poker & get the fuck off my phone. The main problem - agents have made me so productive, and given me so much leverage (especially with content) that it’s hard to stop working.
My dream AI agent tool: - Opus 5.5 Level Model (never delegates to worse) - Spins up cloud and local code sessions - Dots level real time voice / tool use - Codex level Computer use on local sessions - Claude Artifacts quality decks, docs and design - Mobile app is UI of duet -
View quoted postCan someone explain what AI plans will look like in 6 months? Who has a measured take on this? What will be the best plan for 200 or even 500$/ month in 6-12 months?
it's gonna be hilarious when everyone looks back 6 months from now and remembers getting $5K in inference for $200 a month...
View quoted postok @TellaHQ MCP is quite good now.
Claude duet will have local Mac access! And apparently they only use Opus 5.5. All good news.
@rileybrown Really appreciate the feedback. We're working on remote control and local mac access. More to come on that soon! We only use Opus 5.5, so I'd love to hear more about what feels like "Opus 5.5 lite" - examples would help us a lot!
View quoted postOkay to make it easier I have condensed all my Claude Code Projects into one project and named it Duet.
Claude Code Projects is Duet except you get to pick the model and you can specify if you want to run a Claude code session on your computer. It’s coordinator that can do ANYTHING. And it’s easy to do things in parallel. Claude Code projects is the best feature in the Claude
This is sick actually. You guys should stream this somehow.
Sequoia is running an experiment We’re hosting a poker bot tournament, with rules for the age of AI There will be prizes, an incredible guest list (ex @polynoamial), and some plot twists The brainchild of my partner @abhishekm1636 Excited for this!
View quoted postYes… Obviously! AI is an amplifier. I think 300 people have DMd me about video editing with Opus. “What skill should I use?” And I always tell them. Don’t “use” a skill “Learn a skill”… Take 20 hours over 2 weekends to learn premiere pro / davinci resolve or CapCut. Make a video manually. Learn the vocab. Learn pacing. Learn jumpcuts. Learn to make a good video. Then you’ll be able to take the expressions you’ve learned and turn it into repeatable workflows.
Unpopular opinion: the people getting the most out of AI are the ones who could do the work without it.
View quoted postClaude Code Projects is Duet except you get to pick the model and you can specify if you want to run a Claude code session on your computer. It’s coordinator that can do ANYTHING. And it’s easy to do things in parallel. Claude Code projects is the best feature in the Claude App. Very underrated. Claude Duet is clearly not using Opus 5.5 for all requests. And once you go Opus 5.5 anything worse is intolerable.
I've churned on Claude Duet and i'm back using claude code projects.
RT Adam Saks Helpful and accurate summary if you’re wanting to get caught up to speed on state of (mainstream) personal agents.
My unfiltered thoughts on personal agents. Grokbot has the best UI and it’s arguably the best overall. In a lot of ways they’re ahead. Their desktop app feels the best, and their multi-bot coordination is great: you can set a team lead and it passes tasks to the other bots.
View quoted postAnthropic S-1 Slide Deck Visualized Created on Claude Duet Here's the link to the deck https://claude.ai/artifact/NmSxR91LjbDjTFMPC8Cosz#slide-1f0034895dc2
Anthropic has filed an S-1 registration statement with the SEC related to its proposed IPO. Read more: https://www.anthropic.com/news/s1-sec Legal disclaimer: A registration statement relating to these securities has been filed with the SEC but has not yet become effective. These securities
View quoted postis there an api for a notebooklm style podcast? would love to be able to do this from any agent platform.
Yup, make it free, monetize in other ways.
My take: SaaS is dead. Software is charity now, make money elsewhere. No one can make $$ selling the same features for years anymore. Instead: - Use good software to collect user data and sell data. - Use good software to deliver outcome and sell outcome. - Arbitrage
View quoted post