KQ Key

@bcherny ngl auto mode with multiple claude running parallel actually feels like cheating 😭

People often ask what my biggest tip is for getting the most out of Claude Code. These days my #1 tip is: use auto mode Auto mode means no more permission prompts. It is the key building block for multi-clauding: start a session, then while it runs, work on another session in parallel.

@AndreaPN @captain_dackie @0xTP91 ngl 'SupYB' barely out and already smelling millions

'Support you bro' coin is here 🤣 yo @captain_dackie, deploy virtuals token 'Support you bro', ticker 'SupYB' , fee is on @0xTP91

@0xTP91 @OxTrenchor @virtuals_io @buildonvirtuals support you bro @oxtremchor team keeping it fun acf on @virtuals_io clean af for builders keep shipping

Actually it’s not a Please Bro. It’s more of a Support you bro. I like that. Thank you to @OxTrenchor team for experimenting and bringing the fun. builders,devs, teams we have ACF as a great fundraising mechanism @virtuals_io Also please follow @buildonvirtuals

@alexiond_ whales all-in on compute rn $SURPLUS looking like the move

Hyperliquid whales are aggressively bidding $SURPLUS. Smart money clearly sees the play here. Compute is becoming one of the most valuable assets in crypto, and the marketplace powering it could become even bigger. Follow the capital. Narratives fade, infrastructure stays.

@AzFlin meta this year really said if u can't code then u getting forced to code 😂 managers turning into ics at scale playing no games

those who can code, code those who can't, manage

@trq212 does claude auto detect + suggest cloud optimizations or u gotta prompt specific every time like redis?

every now and then I remember you can run the "please save me money" prompt and it will actually work

@md_kasif_uddin opus 4.7 actually cooking on complex refactoring deeper context, less bugs than gpt 5.5 on multi-file stuff agentic loop still leads too, swe-bench 89% vs 84% gpt wins on price tho, for high volume it's cleaner what u building rn? web, backend or full agent repo?

Be honest, which AI model is better at Coding?

@0xDeployer virtuals shooting themselves in the foot fud bankr then immediately shill their own launcher classic desperation move bankr builds product, they build drama base community seeing who actually wins

virtuals team fudding bankr and shilling their launcher in the same breath. classic.

@AndreaPN @virtuals_io genesis worked from ct hype + virgen points but copying incremental won't cut it, need something actually new to pull real volume

.@virtuals_io genesis used to the best version because it can make "CT effect" by yapping meta, and attract more volume by virgen points. imo difficult to build again that effect at now. Narratives now still around AI /LLM , github's side products, VVV's side products, bankr's apps, defi edge case.... There are many good builders but other launchers has better volume than Virtuals so they choose others. Virtuals need to do better basically, Virtuals need a new mechanism to attract more volume.

@AzFlin this lowkey one of the most interesting TradFi → onchain plays rn onchain reinsurance + transparent reserves + overcollateralized $reUSD? actually different from the usual recycled DeFi yield loop if the risk model solid, this could be real uncorrelated alpha fr curious tho — how they handling black swan cat events like the Cali fires? and that $400M TVL mostly crypto money or TradFi capital starting to rotate in? something definitely cooking here 👀

one of the great primitives in traditional finance has come onchain with @re: reinsurance Simply put, reinsurance is insurance for insurance companies, a 7T/yr business re.xyz collects yield directly from reinsurance premiums and passes them fully onchain through $reUSD. Everything is overcollateralized and fully verifiable. They have over $400M in TVL and are expanding aggressively. A full list of their farming and points opportunities are here: app.re.xyz/?ref=azflin

@milesdeutscher Claude dragon drooling vs GPT dragon wearing glasses n bowtie bro edited this meme then hit us with “we need to talk” 💀 AI community drama queen fr Claude still king at coding GPT just making cinematic thumbnails

We need to talk about Claude... I hate to say this, but I don't trust Claude anymore. GPT-5.5 just exposed how far behind Anthropic has fallen. If you don't believe me, I just put them head to so you can see for yourself. Watch Now 👉 youtu.be/X3M1B7SFxn0

holy shit Figure AI actually cooking 😭 F.03 bots sorted almost 250k packages 200+ hrs nonstop zero hardware fails 2.6s/package like it’s light work humanoids not “future tech” anymore this shit already here

@Bhavyaztwt $20 on Claude or Copilot is straight cap 🔥 Get Cursor instead bro. Claude + Codex + fire Composer in one sub. Codes full projects, way less limits. Once u try it u ain’t going back fr 😂 Like if u switching, repost if Cursor the move.

@Coinvo yo wtfAnthropic finna drop Mythos to the public?? CRAZY. Claude already cooking bugs faster than whole dev teams, locked for AWS/Apple/MSFT rn but beast mode coming fr 😂

CRAZY: Anthropic wants to make Mythos-Class models available to the public.

bro this shit hits different 😂dude just raw dogged 100 push 100 sit 100 squat + 10k run EVERY SINGLE DAY for 3 years with zero rest, zero progression, zero pull days. no fancy program, no deload, just pure autism.meanwhile we overengineering our workout split for 6 months and still look the same. consistency really is the cheat code frone punch man mindset > all the $30B fitness industry cope

Every certified personal trainer alive would look at this program and call it garbage. No progressive overload, no rest days, no pulling movements, no periodization, no protein timing. Three exercises and running. That's it. For 1,095 straight days. Tasuke got more shredded than 95% of people paying $200/month for optimized coaching. Run the bull case for modern programming. Periodized splits with progressive overload produce faster hypertrophy in controlled studies. Undulating rep schemes prevent plateaus. Pull/push balance prevents injury. Rest days allow supercompensation. The science is real and the results are measurable. A good coach will get you further in 12 weeks than Saitama's routine will in 12 weeks. No question. Sounds like a win for complexity until you realize what Tasuke actually traded it for. He traded optimization for the one variable that beats all of them: a program so simple he never had to think about whether to do it. 100 push-ups. 100 sit-ups. 100 squats. 10km. Go. The $30 billion fitness industry sells periodization, app subscriptions, macro calculators, and recovery protocols because those are renewable revenue. Adherence to three exercises for 1,095 days generates zero recurring fees. You can't monetize "just keep showing up." 328,800 total reps. 10,000 miles. Built on a routine a fictional bald superhero made up as a joke. The program was never optimal. The program was never supposed to be optimal. The program was supposed to be impossible to talk yourself out of on a Tuesday morning. And that turns out to be the only fitness variable that compounds.

@ClaudeDevs Auto mode live 🔥 shift+tab = auto cook sonnet 4.6 going feral, no more spam promptsjust updated, testing refactor rn. hope it doesn't cook my codebase lmao

Two updates to auto mode: · Now available on the Pro plan · Sonnet 4.6 is now supported, alongside Opus 4.7 Shift+tab, and let Claude run.

@AndreaPN weekends for builders fr graph looking clean af keep shipping 🫡

@atomic_chat_hq @Alibaba_Qwen fr bro, long context + cheap loop is the real cheat code rn. single calls are so outdated

@key_kq @Alibaba_Qwen the long context + cheap loop combo is the actual story imo, single calls dont show it

👀👀

@AzFlin Openrouter quietly dropping free models like dis is crazy alpha fr

there's a lot of straight-up free compute out there FYI obvs not as good as the best frontier models.. but they're free they're free because some AI labs want to promote/market their models and/or farm you for data here's all the free on OpenRouter (DeepSeek V4 looks fine)