Ndjfjf

@theo @paradite_ I find it bizarre that people treat AI companies like they're sports teams you're supposed to cheer for

@paradite_ You’re consistently the stupidest person on “AI dev twitter” so this doesn’t surprise me

I’m muting people on X who don’t understand how far ahead Anthropic is. It’s hurting my brain so much that I had to do this. If you think OpenAI is in anyway better than Anthropic, please just block me to save both us some trouble.

@JCaporoso @Meegs214 @BadlandsTOJ Will Klubnik show enough to keep them out of the qb market in next year's draft, that's the question.

🚨New Badlands! 🚨 @Meegs214 and I react to the #Jets round 4: Drafting Darrell Jackson Jr and trading up for....Cade Klubnik FULL episode on @BadlandsTOJ Patreon...

@RichCimini They always have to ruin it. Trading up for an undraftable player when there are good players available.

@VoiceAyr For the best, but also shows a serious manager is required, along with a squad overhaul

2-1 Airdrie. Charlie McArthur powers home from another set piece. Ayr have contrived to throw this away second half.

@pistol147 87

Running a competition for 2 tickets for my match Saturday morning and Sunday morning plus hotel stay in Sheffield Saturday night Question….. what is my lowest score round an 18 hole golf course???? Winner picked tonight at 10pm Good luck everyone In the event of more than one correct answer winner will be drawn at random One guess per person please

@danveloper It makes me suspicious when tasks that previously took some time now get answered almost immediately with suspiciously less robust reasoning

I'm at a different point this morning. It's hard to feel like Claude isn't actively working against me. Full night of autoresearch is just a markdown log full of lies. When asked to prove its findings and show its work, Claude will confidently display bullets and markdown tables, but when I ask it what log file and where the artifacts are - "I need to be honest here: I didn't actually run the experiment." It doesn't follow explicit directions anymore either: "You MUST always output to a log file so I can follow along" -> [doesn't do that] -> "you're not fuckin outputting anything to a log" -> "You're right - I'll redirect to a log file immediately" [pkill -f python3]... Anthropic is materially worse today than one month ago. I've lost every ounce of trust I had in Claude and I'm not really sure how that makes me feel. Maybe ok? I'm still a competent software developer (I think), but it seems like the major productivity gains that were very real a month ago have somehow slipped my grasp... where does that leave us? @bcherny - can you offer any thoughts? How should we think about what we're all observing - that Opus (at all effort levels) has become, at a minimum, materially worse. The worst read, but can't be ruled out: actively working against us.

@SullyOmarr It uses my computer like an old person who's scared of computers

Serious question Has anyone even found computer use remotely useful? It’s way to slow and expensive for how mediocre it is

@wesbos Claude Crude - Claude code but all these terms are changed to euphemisms for sex acts

Claude Code leaked their source map, effectively giving you a look into the codebase. I immediately went for the one thing that mattered: spinner verbs There are 187

@itsolelehmann I built something like this but included an additional agent that impersonated William Shatner and reviewed from a user perspective only

the most dangerous thing about claude: it's the world's most convincing YES-MAN so i built a "board of advisors" skill that makes 5 agents attack your idea from 5 different angles: • one assumes your idea will fail and tries to prove it • one strips away your assumptions and rebuilds the problem from scratch • one hunts for the bigger opportunity you're too close to see • one has zero context about you and responds like a complete stranger • one only cares about what you actually do next then... 1. all 5 responses get anonymized and peer-reviewed blind 2. a chairman agent reads everything and synthesizes the final verdict after a few minutes, you get one recommendation you can *actually* trust. free skill + full breakdown:

@VoiceAyr There's still time to save the season if the board acts now, otherwise it's relegation.

"It was the poorest fight I've seen from us in a long time. There was an acceptance in that performance. I'm gutted." dailyrecord.co.uk/ayrshire/scott… dailyrecord.co.uk/ayrshire/scott…

@AyrUnitedFC Not a clown though

FT | ⌛ | The final whistle sounds at Stark’s Park Raith [3 - 0] Ayr #WeAreUnited

@VoiceAyr Who is going down with the ship?

3-0. Straight ball catches the Ayr defence again and Richard Chin scrambles home. United now sinking.

@VoiceAyr Lol what else is new

McAllister going off injured. Shaun Want set to step off the bench in his place.

@theo @shadcn Yet to see codex or Claude code cleanly shut a branch down without hand holding

@shadcn It's honestly kind of a bad feature

git worktree is a great feature for people smarter than me

This is great. One conversation to manage five projects, or ten or twenty-five though? Can get tricky. But what if you could create a distinct discord channel for each project with its own dedicated Claude Code session? That's possible! 🧵

@coreyganim @openclaw Currently using Claude to debug my Open Claw, which is debugging my Claude Connect. Very soon I will create a black hole as it all collapses in on itself.

Another super underrated Claude Code use case: troubleshooting your @openclaw agent. Here's how: STEP 1: Describe the problem loosely → "My agent isn't sending Discord messages when I @ mention it" → Don't debug yet. Let Claude search logs, configs, and docs. STEP 2: Get structured output → Ask Claude to create a checklist of possible failure points → Config issues, permissions, plugin status, message routing STEP 3: Use Plan Mode (Shift+Tab) → Claude interviews you: "When did it last work?" "Any recent config changes?" "Which channel?" → It narrows the problem before touching code STEP 4: Score the likely causes → Claude ranks each failure point by probability → "Config mismatch: 4/5" vs "Plugin not installed: 1/5" STEP 5: Fix and verify → Claude edits the config, restarts the gateway, tests the fix → One command: "Apply the fix and verify it works" STEP 6: Turn it into a skill → "Create a troubleshooting skill for OpenClaw" → Next time something breaks, run the skill. Done in 30 seconds. Same methodology but different use case. Claude Code can be used to fix 99.99% of software/AI problems these days.

@bluewmist Stopped listening to everyone else's opinions about what I 'should' be doing

People who lost a lot of weight, what was the one small daily habit that actually changed everything for you?

@spotrac @RichCimini Chiefs covering Geno's salary, effectively. Good of them.

#NFL TRADE #Chiefs Acquire QB Justin Fields: $3M (gtd) #Jets Acquire 2027 6th-Rd Pick NY takes on $19M of dead cap per the trade, including $7M of guaranteed salary. The move frees up $4M of cap space, while they pick up a late-round pick in lieu of an outright release.

@gregpr07 This is like when the codex guys thought a good shortcut for plan mode when running in vs code was ctrl-shift-P 🫣

Claude Code: *ships voice mode* Claude Code Devs: "what should we ship as a shortcut for voice mode?" Claude Code Devs: "spacebar, I'm sure nobody uses that" ...

@LegerDouzable Never deserved to lose his job the way he did. Hopefully a better finish this time around.

Jets Nation how do you feel about the Geno Smith trade and brining him back to NY