GatorXman

@Jason I’m in

We started an AI founder twitter group... reply with "I'm in" if you're a founder and want to be added

@thsottiaux Windows support. I shouldn’t need a custom AGENTS.md to make PS commands functional.

Hello builders. What are we getting wrong with Codex, what can we improve?

@amorriscode @lydiahallie Is it actually functional in Windows? Or does the windows user industry get some ghetto potato

Today we're launching a rebuilt version of Claude Code on desktop. The app has been redesigned for the ground up to make it easier than ever to parallelize work with Claude. I haven't opened an IDE or terminal in weeks. Excited for you all to give it a shot!

@0xrwu Disable adaptive thinking. That fixes it.

I swear to god Opus 4.6 went from 1 month ago = 110 IQ 2 weeks ago = 90 IQ a few days ago = 50 IQ It can barely implement anything w/o bugs! What is going on?!?!

@claudeai Is it still a potato on windows machines??

We've redesigned Claude Code on desktop. You can now run multiple Claude sessions side by side from one window, with a new sidebar to manage them all.

@KillaXBT Correct. During the liquidity crunch that hits markets at the end of this month. Next leg down will hurt.

Just cracked the $BTC bottom. Whether you like it or not, we are going sub 60K an there is absolutely nothing you can do about it.

@lydiahallie What about the lobotomy you gave Opus???

We're aware people are hitting usage limits in Claude Code way faster than expected. Actively investigating, will share more when we have an update!

@MoeCanDoIt It 💯 got nerfed. The change was sudden and massive. Opus can’t even do simple tasks anymore. Riding this out in Codex for the time being.

I'm on Claude's $200 plan, and for some reason it's getting dumber. I'm talking simple contract review stuff where it COMPLETELY misses important details. Did it get nerfed? What's happening?

@SucciMas @MoeCanDoIt It’s definitely not perception this time. Performance suddenly fell of a cliff. Last week it was even refusing or incapable of doing simple black and white task like “show me the exact system and user prompt you see for that API call” 80k context into a convo.

@MoeCanDoIt Perception of 'dumber' is almost never model regression. It's the same model on a different day of your context window

@xSoloTrades @mkurman88 I thought I was tripping. Same thing. Genius level past few days. Today can’t even follow black and white directions.

@mkurman88 Bro, today Claude is a literal potato. I've been deep in the trenches using Claude Code like 14 hours a day for the last month with such great success and today it can't perform the simplest things

I just canceled my Claude subscription. It couldn't resolve a straightforward issue for 3 hours, even though I pointed out several times where the bug lies (I was afk, so I couldn't do it by myself, but that's not the point). I asked Codex to fix it -> 1 minute and done. I asked Codex to review other things Claude made - with GSD, 20 steps, discussion, plan with research, execution. It has been working since Saturday. And what? Everything was faked. I don't get it...

@thismacapital Funny I’ve had this setup for a while. But I use a separate model (Codex) as my judge. Always a fresh session so there’s no contamination.

TL;DR : Anthropic a développé une architecture 3 agents pour faire coder Claude en autonomie pendant des heures sans perdre en qualité : Le problème : • Claude perd le fil sur les tâches longues (context window saturée) • Claude s'auto-évalue trop positivement : il dit "c'est bien" même quand c'est moyen • Le design frontend généré est souvent générique La solution : 3 agents séparés : 1. Planner : découpe le projet en tâches 2. Generator : code les features une par une 3. Evaluator : juge le travail du generator (inspiré des GANs) Deux insights clés : • Context resets > compaction : au lieu de résumer le contexte (compaction), ils reset complètement l'agent et passent un artefact structuré au suivant. Ça élimine le "context anxiety" où Claude bâcle la fin parce qu'il croit manquer de place. • Séparer le juge du créateur : un agent qui évalue son propre travail est toujours trop indulgent. Un agent séparé, calibré pour être critique, donne un feedback honnête. Le generator itère 5-15 fois en réponse aux critiques. Pour le design frontend : • 4 critères de notation : qualité design, originalité, craft, fonctionnalité • L'évaluateur utilise Playwright (MCP) pour naviguer sur la page avant de noter • Résultat : des designs beaucoup plus originaux et moins "template AI" Pour le code long-running : • Sessions de plusieurs heures en autonomie totale • Le planner crée des specs détaillées, le generator code, l'evaluator valide En gros : fais apparaître un deuxième agent pour critiquer et challenger le travail du premier

@bcherny Am I the only 100% YOLO user out there?? Even during planning I bypass ☠️

no 👏 more 👏 permission prompts 👏

@Notnate2188 TY for your service and exact message delivery we are all thinking

In response to Dan Bongino. Also, we were not screaming at him and we did not get kicked out. He ran away with his security.

@SenFettermanPA Fag

I don’t follow the polls. As a Democrat, I unapologetically stand on the side of moral clarity and our ally Israel.

@UF You make it embarrassing to be a gator.

The University of Florida was recently informed by the Florida Federation of College Republicans (FFCR) that it has disbanded the local chapter of College Republicans (the Local CR).  This request is based on the FFCR’s findings that some Local CR members engaged in a pattern of conduct that violated its rules and values, including a recent antisemitic gesture. The FFCR also requested that the university deactivate the Local CR as a registered student organization while it looks to reorganize and pursue reinstatement under new student leadership. The University of Florida has emphatically supported its Jewish community and remains committed to preventing and addressing antisemitism and other forms of discrimination and harassment that are threatening and disruptive to our students and to the teaching, research and expressive activities of the campus community. The university also supports the rights of organizations, such as the FFCR, to take decisive action in addressing conduct that is antithetical to its principles. In compliance with its policies, the University of Florida is in the process of deactivating the Local CR as a registered student organization.  When the FFCR is ready, the university will also assist it with reactivating the Local CR under new student leadership.

@PressSec Shut up ho

This story is 100% FAKE NEWS.  CNN decided to run this garbage based on three anonymous “sources familiar with discussions.”   This is despite the fact that myself, the Secretary of War, the Secretary of State, and multiple lawmakers (who were actually present for the recent classified briefing) have directly disputed this false reporting.  THE TRUTH: The Pentagon has been planning for Iran’s desperate and reckless closure of the Strait of Hormuz for DECADES, and it has been part of the Trump Administration’s planning well before Operation Epic Fury was ever launched.  The idea that Chairman Cain and Secretary Hegseth weren’t prepared for this possibility is PREPOSTEROUS. The President was fully briefed on it, and a goal of the Operation itself, to annihilate the terrorist Iranian regime’s navy, missiles, drone production infrastructure, and other threat capabilities is quite literally intended to deprive them of their ability to close the Strait. President Trump will not allow rogue Iranian terrorists to stop the freedom of navigation and the free flow of energy. Wiping out these terrorists who indiscriminately target civilians and attempt to hold the global economy hostage is part of the ongoing noble U.S. mission. The Fake News is working overtime to discredit President Trump, his Administration, and our U.S. Military, all of whom are working 24/7 to eliminate the threat of the Iranian regime. It’s a complete disgrace to witness this from the media, and we will keep fighting back against it.

@LindseyGrahamSC Stfu phaggot

I was just informed that the Spanish government has permanently recalled their ambassador to Israel. This is hard for me to absorb. Spain is a member of NATO, and the United States and Israel are in joint operations against the Iranian regime who openly calls for the destruction of the Jewish State, attacks against the West, and seeks to purify Islam in its own image. The religious Nazi regime in Iran is the problem, not the Jewish State. I hope Spain’s actions will not encourage the tyrannical, fanatical regime in Iran — that abuses its own people — to hang on. Time will tell. reuters.com/world/middle-e…

@ezcontra Good luck fighting that liquidity tidal wave that’s about to hit

Down only until 200d

Kinda think this is the exit pump on stocks. Will be looking for shorts on es/spy today into next week. 200d gets tagged next month

@BillAckman fuck off dude. you literally pushed for this. go back to your motherland

The greatest sacrifice one can make is to give up one’s life for another, in this case for hundreds of millions of Americans, 90 million freedom-loving Iranians, and the rest of the world from whom a nuclear and terrorist threat is in the process of being removed. Words are inadequate to honor those who made the ultimate sacrifice.

@ByronDonalds @realDonaldTrump Dance ✡️ puppet dance. Good boy that’s how daddy Bibi likes it.

President @realDonaldTrump has acted decisively against Iran's nuclear threat and terrorism. Our troops are now executing Operation Epic Fury. Erika and I pray for their safety and success. Peace is achieved through strength, not weakness. I stand with President Trump, our military, and the hope this leads to lasting security in the Middle East. God bless our troops. God bless America.