AILearning

Just on here to soak up info & learn more about AI and various AI tools, etc. to utilize in various work & personal workflows.

[AILearning @A_I_Learning]

Those who believe in the Bible can appreciate this one.😃

[AILearning @A_I_Learning]

@larsencc My boss owns the universe. I win.

[AILearning @A_I_Learning]

@pradeepXkapoor Opus 5.5 still makes more dumb choices than Fable 5.1. Opus 5.5 is very very good ... usually. But today for example it did a bunch of stuff I never gave it permission to do and went nuts in multiple repos taking liberties I never said it could. Fable has never done this.

[AILearning @A_I_Learning]

@claudeai I thought I was seeing a glitch with the --% now I know its real there is no limit on 5 hours!! Woohoo! Good now I can clean up the mess that Opus 5.5 made today without my permission quicker. It totally ruined my evening and is stealing my sleep.

[AILearning @A_I_Learning]

No more new chat IDs! /fork-archive-trim-continue forks chat when context is high to save transcript in another chat ID, archives it to clear it from the main account & separate into archive (can be read), trims the chat ID, writes a brief, trims & starts same chat from scratch.

[Nick Rempel @nbrempel]

I've started calling Claude big dawg and I think improves performance

[AILearning @A_I_Learning]

@nbrempel 🤣

[AILearning @A_I_Learning]

@trq212 I don’t know what I feel because I’m too tired from being up all night on Claude code giving it all of my decisions so that it will run during the day when I’m gone.

[Sauers @Sauers_]

One of the subagents just made a figure so bad that the lead Claude immediately killed them and replaced them with a new agent.

[AILearning @A_I_Learning]

See how dangerous these things are! Spilled coffee and everything! Can’t take it anymore! 🤣🥱

[AILearning @A_I_Learning]

I’m not leaving @Anthropic, because they never hired me. But if I was leaving, it would be because @claudeai @claude_code caused me to spill my coffee this morning. See the pictures for evidence. Claude admits it openly. 💯☕️

[AILearning @A_I_Learning]

TRUE STORY!

[AILearning @A_I_Learning]

I’m not leaving @Anthropic, because they never hired me. But if I was leaving, it would be because @claudeai @claude_code caused me to spill my coffee this morning. See the pictures for evidence. Claude admits it openly. 💯☕️

[AILearning @A_I_Learning]

I’m not leaving @Anthropic, because they never hired me. But if I was leaving, it would be because @claudeai @claude_code caused me to spill my coffee this morning. See the pictures for evidence. Claude admits it openly. 💯☕️

[AILearning @A_I_Learning]

2/2 I asked Claude why it never ends, never gets to a stop, it said "It never reaches 100% because the list grows about as fast as it shrinks. Every finished item gets a common-sense check, and each check adds new items. Today 2 items were finished and 5 new ones were added."

[AILearning @A_I_Learning]

1/2 My /common-sense skill must be genius because Claude Code never gets to 100% done because it keeps coming up with new coding ideas automatically without me needing to think of what else my Markdown Studio app needs! Read my reply to this for next thread.

[AILearning @A_I_Learning]

@emmjay_init Once I started using it, I got better results in everything. It took 15 iterations to get there. About half of those driven by Claude about the other half by me once it hit a brick wall. And I have a lessons learned.

[AILearning @A_I_Learning]

I just had this idea of making a website where in order to use it you had to prove that you’re AI instead of prove that you’re human. LOL.

[AILearning @A_I_Learning]

@emmjay_init And he eats for a day. Teach a man to fish and he…

[AILearning @A_I_Learning]

@emmjay_init So all I have to say is ‘use/apply the style guide” and it knows what to do.

[AILearning @A_I_Learning]

@emmjay_init What do you mean? Prompt for what?

[emm jay (glowie) @emmjay_init]

@A_I_Learning Yeahhh, I don’t believe in ai consciousness either 😬

[AILearning @A_I_Learning]

@thenikacode Yeah, well you can’t interrupt it in the middle of its work all the time unless it’s critical. That’s true, even with my skill. Eventually, it’ll get it so it’s not a big deal.

[AILearning @A_I_Learning]

I am building a roles-based Claude code framework to mimic how humans work in an org & run my normal Claude set-up. Today one session refused to help another because it said ‘I only do XYZ.’ This is similar to how humans say ‘Sorry I can’t help you. That’s not my job.’ Success!