Rich Bira

The Generative UI is starting to take shape, but do product teams understand the implications? Also, @AnthropicAI's approach is different than @Google's A2UI They are both working towards the same end: rendering in the UI. I guess middleware like @CopilotKit will be needed

@GaelBreton @karpathy 🧐hmm. You have a point. lol

@karpathy All of this just to get Mythos access.

Personal update: I've joined Anthropic. I think the next few years at the frontier of LLMs will be especially formative. I am very excited to join the team here and get back to R&D. I remain deeply passionate about education and plan to resume my work on it in time.

@DannyLimanseta @karpathy This was not on my bingo card today!

@karpathy wow. I did not see this coming. Congrats! I look forward to seeing what you will push forward with the Antrophic team.

So big! Congrats @karpathy

@TommiPedruzzi Claude

I just finished creating my most valuable PDF yet: "18 Claude Cowork Workflows for the Entire eBook Business" (44 pages). I might charge for this in the future, but for now... Reply "Claude" and I’ll DM it to you for free (must follow)

@Jacobsklug OS

I built a Claude Cowork OS that replaces OpenClaw and runs on autopilot. Manages my business & personal life tasks. I created the whole playbook so you can re-build it tonight. What's inside: • The exact foundation prompt • 3 level orchestration map • Memory template for global context • Routing table for file management • Starter workstations (finance, content, community, habits) • Project file structure • Single prompt that builds the entire folder tree Follow + Comment 'OS' and follow. I'll DM it to you.

@codyschneider FBmanager

last Friday I deployed a Facebook ads agent for a startup and over the weekend it optimized itself from $17 phone number leads into $3 leads this is the real GTM engineering agents in the wild and I made a Notion document and a .md skill file so you can do this exact thing too it includes: 1. How to make on brand ads with nano banana 2 2. How to upload these to facebook ads via the API 3. Have an agent manage the ad account based on live data from the account via data pipeline + data warehouse Outcomes we deployed an AI agent for a startup last Friday to manage their facebook ads account day 1: the cost per phone number lead was $17 over the weekend this agent made and published 30 new pieces of ad creative, optimized the ad account, and reduced the cost per action Day 4: the cost per phone number lead was $3 this is a virtual employee that's working full time. like this post and comment "FBmanager" and I'll send you the Notion file and MD file

@itsalexvacca ADS

We built 12 Claude Code skills that run our entire paid media ops across Google, Meta, and LinkedIn at ColdIQ (and we're giving the whole pack away). Our head of growth Ivan Falco runs $200K/month in ad spend from a terminal. It's how we doubled client load this year without losing quality. The skills do the work that used to fill our media buyers' calendars: spot creative fatigue, adjust bids, upload audiences, run bulk edits, flag broken campaigns, build reports. Each skill does a specific job: Google Ads: → keyword-analyzer: audits quality scores and finds keyword gaps → negative-keywords: reviews search terms and blocks wasted spend → performance-auditor: compares periods and flags what changed → search-terms: surfaces queries burning budget with zero conversions Meta Ads: → audience-builder: turns CRM lists into custom audiences → creative-fatigue-analyzer: spots declining CTR before the metrics flag it → fatigue-monitor: flags when your audience is saturated → spend-tracker: tracks budget pacing across every campaign LinkedIn Ads: → audience-builder: builds targeting audiences at scale → bid-optimizer: adjusts bids across campaigns in bulk → bulk-editor: mass edits campaigns, ads, and naming in seconds → creative-builder: generates ad creatives from brand specs You drop them into Claude Code, connect your ad accounts, and tell it what you need. It reads the skill, plugs into the platform, executes. 300+ hours of work went into building these. Comment ADS and we'll send all 12 over.

@codyschneider AEOdashboard

my team didn't want me to give this away for free. But I'm going to do it anyway it's the SEO & AI search dashboard I built in Claude Code it connects to your Google Analytics (GA4) and Google Search Console and Claude Code builds it in 5 minutes and I made a Notion document and a skill file so you can build this in Claude Code yourself in literally minutes the dashboard has three tabs: 1. AI Search - How much traffic is coming from ChatGPT, Perplexity, and Gemini ETC. It aggregates the GA4 data and gives single number 2. Paid ads - which keywords rank top 3 for but still pay for ads on, you should cut these to save budget 3. Organic overview - sessions, conversions, top landing pages, demographics. The single view for what is working I built this because this is how I drive our SEO and AEO forward it gives me the insights I need to allocate budget and prioritize what content to work on next I decided to give it away because most companies have no idea AI search is already sending them traffic like this post and comment "AEOdashboard" and I'll send it over

@OverlyTrev Check out my reply to this. Spoiler, Chicago needs it. x.com/RichAboutRE/st…

Take an escalator or elevator right from your building to a personal autonomous vehicle that will take you straight to the airport or a popular destination (Lower Broadway in this case) in less than 10 minutes. This is the future of travel. The Boring Company and autonomous Teslas go together like peanut butter and jelly.

@Nicolascole77 social

I just packaged 102 of my best writing templates into 5 Claude Skills. This is everything you need to: • Write scroll-stopping hooks • Create X content for yourself • Ghostwrite content for high-paying clients Comment "social" and I'll send it across ASAP (for free).

@boringcompany @ChicagosMayor @nbcchicago @ChicagoBears @ArlingtonHtsGov @elonmusk With today’s Nashville Music City Loop announcement delivering ~9-minute airport access, any renewed interest in reviving Chicago’s 2018 Express Loop? That proposal was officially benchmarked at ~12 minutes one-way from downtown (Block 37) to O’Hare. Privately funded Tesla pods that would beat existing options 3-4×. Vegas proved the tech works. Nashville is building it. As @nbcchicago’s recent “Superstation to Nowhere” investigation showed, Chicago already spent $400M+ on the unfinished Block 37 superstation shell. It’s still sitting idle and mothballed. What a great way to turn that taxpayer investment into a revenue-generating asset, attract business and travelers back to downtown, and help tackle growing Loop vacancies, all with zero new public dollars. And if the @ChicagoBears really leave Soldier Field, let’s hope they land in Arlington Heights. That would open the door to extend the tunnel from O’Hare straight to the proposed Bears Entertainment complex. If Chicago doesn’t move on this, who knows.. maybe @ArlingtonHtsGov will. Either way, it would cut Bears game-day traffic and boost visitors to the attractions at the complex. Chicago or the region could be next! 🚀 #ChicagoLoop #BearsArlington PS. Remember this post x.com/chicagotribune… Thanks, @grok for the real-time research and fact-checking. I kept adding the ideas (Block 37 current status, station costs, vacancies, revenue upside, Bears extension, etc.) and Grok validated the details. Exactly how fast human + AI teamwork should work.

TBC is excited to announce the first residential Music City Loop station agreement, serving those living at the beautiful Prime, Alcove, and Paramount towers in downtown Nashville. Residents will take an elevator directly to a Loop station within the building, board a Tesla, and be at the airport in ~9 minutes or Lower Broadway in ~2 minutes.

Thanks Meta4, I always appreciate a quick reality check from you. I did a quick read-up on this. Basically, my understanding is that Titans literally memorizes context into its parameters during the session. Because memory is baked into the weights rather than stacked in a KV cache, compute scales perfectly linearly, correct? Anyway, back to our regularly scheduled programming: wiring up Gemma 4, Embeddings 2, and Gemini 3.1 Flash Lite. Having a lot of fun building with these right now.

@RichAboutRE @alex_whedon @sierracatalina Happy to chat about it.. but spoiler: the claims here are bigger than reality IMO. At best they get to n*log(n) where titans scales at linearly at N and is already out there.. This paper is claiming victory against a 2017 reality vs the 2024 reality..

Introducing SubQ - a major breakthrough in LLM intelligence. It is the first model built on a fully sub-quadratic sparse-attention architecture (SSA), And the first frontier model with a 12 million token context window which is: - 52x faster than FlashAttention at 1MM tokens - Less than 5% the cost of Opus Transformer-based LLMs waste compute by processing every possible relationship between words (standard attention). Only a small fraction actually matter. @subquadratic finds and focuses only on the ones that do. That's nearly 1,000x less compute and a new way for LLMs to scale.

@dan__rosenthal OS

We scaled from 0→7+ figures in a couple months. Our next goal... - $30M/y in revenue - < 80 person headcount To do this, we’re building out our ENTIRE Company OS on GitHub. And how that's powering ops through Claude Code. Tons of people asked me how to set this up themselves. So I'm making it available to the public. What's inside: 1) Company OS blueprint with full folder structure and .md guides 2) GTM Engineering plugin with pre-built workflows 3) 5 ready-to-use GTM skills (covers everything from outbound copy to ICP modeling to discovery prep) 4) Notion guide on getting started with Cursor 5) The exact file architecture we use across all client projects Want in? Reply “OS” and I'll send you the link. (MUST BE FOLLOWING)

@learnwithella SEO

I built a Claude Code SEO agent that quietly replaces your $200/month Ahrefs subscription 🤯 One prompt → it finds keyword gaps, analyzes competitors, writes content in your brand voice, and tracks rankings weekly All inside Claude Code Perfect for DTC brands and agencies who know SEO matters — but never have time to execute consistently If your SEO workflow looks like this: Log into Ahrefs once a month → export CSV → skim for 5 minutes → close tab → “I’ll write next week” → never do… This agent fixes that 👇 → Connects to Google Search Console and pulls real ranking data → Finds your “gap zone” (keywords sitting at positions 5–20, one article away from page 1) → Uses Apify to break down exactly why competitors outrank you → Learns your brand voice once — then applies it to everything → Writes content that actually ranks (not generic AI fluff) → Tracks rankings weekly + feeds insights back into the next cycle → Optimizes product listings for AI search across ChatGPT, Gemini, and Perplexity No tools you forget to open No generic content No wasted effort Just a system that runs the full SEO loop for you What you get: • Keyword cards with clear action steps • Competitor breakdowns with exact fixes • Weekly content plan from real data • A persistent brand voice profile • AI-search optimized product listings (the new SEO most people are ignoring) Built entirely in Claude Code using Google Search Console I’ve put together the full playbook — prompts, workflows, brand interview, everything Want it for free? Like this post Comment “SEO” I’ll send it over 👇

@AlfieJCarter SKILL

I just built a Claude Code skill that writes, schedules, and publishes your LinkedIn posts every week while you sleep. Feed it your profile, your hooks, and a topic or script → it studies your voice and your best-performing formats → generates three post drafts matched to the right photo and scheduled to LinkedIn automatically. All inside Claude Code. Perfect for GTM engineers and founders who are still writing posts from scratch every week, spending an hour editing AI drafts that sound nothing like them, and manually uploading content to LinkedIn one post at a time. If you're posting on LinkedIn in 2026, you already know the math - the accounts that build pipeline aren't the ones with the cleverest copy, they're the ones posting consistently in a voice that sounds genuinely human. Most people ship two posts a week if they're lucky. This skill solves it: → Drop your profile.md with personal stories, professional context, and voice rules into the skill so Claude writes like you from session one → The skill reads your hooks.md file of proven formats and no-go openers so Claude never produces the lines saturating your space → Pulls from your approved drafts, YouTube scripts, or plain language topic descriptions to generate three options per post → Matches each post to the right photo by reading filenames so every visual fits the post's feeling without manual selection → Fires directly to LinkedIn via Playwright MCP - pastes, uploads, resizes, and schedules to your preferred time automatically → Drops a full week of scheduled posts into LinkedIn from one Sunday morning session No briefing a ghostwriter. No editing AI slop for an hour. No manually uploading posts one at a time. What you get: - profile.md setup guide so Claude learns your voice, stories, and professional context permanently - hooks.md file structure with proven formats and a no-go list built for your specific space - Full skill file that generates, reviews, and approves three drafts per topic in seconds - Playwright MCP setup so Claude publishes and schedules directly to LinkedIn without you touching the platform - One skill you install once and run every week forever Built 100% in Claude Code. I put together a full playbook with the skill file, the profile.md template, the hooks.md structure, and the exact Playwright MCP setup to get this publishing on a schedule. Want it for free? > Like this post > Comment "SKILL" And I'll send it over (must be following so I can DM)

@OverlyTrev Wow, I'm blinded! Congrats!!!

Today was a big day for me ❤️

@learnwithella SEO

I just built a Claude Code SEO agent that replaces your $200/mo. Ahrefs subscription 🤯 One prompt → keyword gaps found, competitors analyzed, content written in your brand voice, rankings tracked weekly. All inside Claude Code. Perfect for DTC brands and agencies who know SEO matters but never have the bandwidth to actually do it consistently. If your SEO workflow looks like this — log into Ahrefs once a month, export a CSV, skim it for 5 minutes, close the tab, tell yourself you'll write that blog post next week, never do... This agent runs the entire loop for you: → Connects to Google Search Console and pulls your real ranking data → Finds your "gap zone" — keywords sitting at positions 5-20, one article away from page 1 → Uses Apify to scrape who's outranking you and breaks down exactly why they're winning → Interviews you once about your brand, customers, and positioning — then never asks again → Writes content in your voice — not generic AI slop that tanks after 90 days → Tracks rankings weekly and feeds what's working back into the next cycle → Optimizes your product listings for AI shopping — so you show up when someone asks ChatGPT for a recommendation, not just Google No $200/month tools you open once and forget. No freelancers writing content that sounds like everyone else. No manually checking rankings and forgetting to act on it. What you get: - Keyword cards with a specific action recommendation for each gap zone opportunity - A competitive breakdown — who's beating you and the exact fix for each keyword - A weekly content plan generated from your real GSC data - A brand voice profile Claude uses for every article it writes - Product listing optimization for AI shopping (ChatGPT, Gemini, Perplexity) — the new SEO nobody's doing yet Built 100% in Claude Code with Google Search Console. I put together a full playbook with the skill files, brand interview, and the exact weekly workflow. Want it for free? > Like this post > Comment "SEO" And I'll send it over (must be following @learnwithella so I can DM)

@itsalexvacca N8N

We run 13 n8n workflows across ColdIQ's entire content, ads, and outbound engine, and I'm giving them all away. Our monthly n8n bill: $384. The same workloads on Claude would cost $60K. That's why I'm not buying the "Claude killed n8n" take. Claude Routines are good at scheduled agentic tasks that need reasoning. They're not the same layer as n8n. They're not a replacement for production GTM infrastructure. Our n8n stack fires 2,000+ executions daily across 13 workflows. Our Phone Finder alone has 41 nodes and waterfalls across 5 data providers. Here's what's in the doc: → GTM Flywheel (81 nodes): domain in, full ICP, lookalikes, prospects, and a tailored content/ads/outbound strategy sent to your inbox → Phone Finder (41 nodes): name, LinkedIn URL, or domain in; Prospeo, FullEnrich, and more waterfalled; verified number in seconds → AI Agent Reply Manager (13 nodes): classifies a cold email reply on Instantly, drafts a response in Slack, waits for your approval before it goes out → Lookalike Finder (35 nodes): domain in, similar businesses by industry, size, and tech signals out → Viral Content Browser (10 nodes): pulls viral LinkedIn posts via Serper, filters by engagement, stores the best in Notion → Feeling Tracker (36 nodes, coming soon): sentiment analysis on any tool across X, Reddit, YouTube, and LinkedIn Plus 7 more covering content, GTM, and ops. Shoutout to Sacha Martinot who built the most complex ones. Reply "N8N" and I'll send you the full doc. Must be following.

@garrytan Me reading this at 2am.. smiling

I came back to code because AI made it possible for me to build at a level I couldn't before. I'm not coding despite being CEO of YC. I'm coding because this is the most important technological shift since the internet and I'd be an idiot to experience it from the bleachers. I'm 45, running the most important startup institution in the world, and I can ship production software at 2am. That's not a distraction from the job. That is the job understood correctly.