Ian

@DumbSMMwtf will become a unicorn

🚨BREAKING🚨 crypto bros invented employment

Most founders think AI gets smarter when models improve. I think AI gets smarter when context compounds. A mediocre model with 2 years of context about your business is often more useful than a frontier model starting from zero. Dreaming feels like the first serious step toward compound intelligence.

@kimmonismus Not “we are close to self-improving AI”, but “we are close to IPO”

Holy moly, Anthropic is getting very serious about recursive self-improvement! One word: acceleration. Insane blog article. Tl;dr: •We are close to an AI capable of fully autonomously designing and building its own successor •They stress this isn’t here yet and isn’t inevitable, but could arrive sooner than most institutions are ready for •Anthropic engineers now ship on average 8x as much code per quarter as they did in 2021–2025 •Task length AI can reliably complete is doubling roughly every 4 months (up from every 7 months) •Opus 3 (Mar 2024) handled ~4-minute tasks; Sonnet 3.7 (a year later) ~90-minute tasks; Opus 4.6 (a year after that) 12-hour tasks •SWE-bench went from low single digits to saturated in two years; CORE-bench (research reproduction) went ~20% to saturated in 15 months •METR found Claude Mythos Preview could work “at least” 16 hours, at the top of what they can currently measure •As of May 2026, Claude authored 80%+ of code merged into Anthropic’s codebase (low single digits before Claude Code launched in Feb 2025) •A March 2026 poll of 130 research staff: median respondent estimated ~4x output with Mythos Preview •One April 2026 example: Claude shipped 800+ fixes cutting a class of API errors 1,000x, work an engineer estimated would have taken a human four years •Claude-written code quality: worse than human in late 2025, roughly at parity now, expected to be strictly better within the year •On the hardest open-ended tasks, Claude’s success rate hit 76% in May 2026, up 50 points in six months •Code-speedup test: Opus 4 averaged ~3x speedup (May 2025), Mythos Preview ~52x (April 2026); a skilled human needs 4–8 hours to hit 4x •In an AI-safety research project, Claude agents recovered 97% of a performance gap (vs ~23% for two human researchers in a week), over 800 compute-hours and ~$18K •On picking the better “next step” in research sessions, the best model beat the human choice 51% (Nov 2025, Opus 4.5) rising to 64% (April 2026, Mythos Preview) •Human comparative advantage, for now: research taste and judgment, i.e. choosing which problems matter and when an approach is a dead end Three possible futures •The trend stalls (S-curve), but today’s capabilities still diffuse widely; they consider this least likely •Compounding efficiency gains, with humans still setting direction; 100-person firms doing the work of 10,000+; they think this is the likely path •Full recursive self-improvement, where AI builds its successors and pace is set by compute; the alignment outcome here is what they’re least certain about

sounds good! now I need only unlimited tokens to process this shit

NO cheating, post your last saved image without any context 💀💀

@Mostafakhafagyy thx for the feedback! yeah,sure! 🤝

Took a quick look. The strongest part is product-aware content, not generic AI captions. That’s the right angle. One small suggestion: show the first 60 seconds of what happens after pasting the URL. Rooting for this. Let’s build and share the progress, experiments, and messes together.

Hey founders ! Looking to connect with people building in: SaaS 🚀 Tech 📲 Automation 🧠 AI tools 📱 Product Development 🔥 Web APP 💻 Devs Drop what you're working on 👇

@vansh22b technically if someone paid you $114 once in less than one hour, you’ve already reached a $1M ARR

I have a plan. To reach $1M ARR. By the end of this year.

@tanya1866 I always thought there was something gay about people using Claude, but this explains a lot

claude is a man and codex is a woman.

@Shahahahzaman no way!!

@burdukes Damn bro, we got same project

That moment when “works on my machine” becomes “works on every machine.”🥺

@ApplyWiseAi @Mostafakhafagyy before: you spend 2 hours a day researching, drafting, editing and posting content across platforms. after: the agent does that in the background and your routine shrinks to a short review block where you only approve what feels on‑brand

@burdukes @Mostafakhafagyy what does the daily routine look like before vs after the agent runs

@Mostafakhafagyy you just paste your product URL and AI does all the dumb, time‑consuming stuff – searching ideas, creating videos, writing and scheduling posts. you only approve what actually fits your vibe, and as it learns your taste we’ll gradually automate almost all of that grind.

@burdukes "Vibe marketing"😄😄 Closed beta sounds interesting, what part of the routine does it actually take off your plate? Always curious how people are drawing the line on what to automate vs keep human.

@DrClownPhD God of War Wifey

🚨BREAKING: New gameplay footage from God of War: Laufey has just been released. The cooking mechanics look incredible.

it's important to have fun while building your product. we shipped this feature in an hour by reusing our existing content generation infrastructure – and it suddenly made marketing fun again, we couldn't stop creating more and more content

Super excited to apply to @hf0! Looking forward to next steps!

we found out that to make AI good at social media, you first need to make it dumber so we fed it brainrot, reels, hooks, slop, bad captions, good captions, and every cursed format we could find. now it finally speaks internet

@pcshipp Claude code wrapper

Hey devs, if you had unlimited Claude Code usage for 1 month, what would you build?

@Anas_founder .wtf

Founders, which domain extension gives startup vibes instantly? .ai .com .net .io .app .xyz .st