Tootfinder

Why use one code when lots and lots and lots and lots of code do trick? #agenticai #ai

I read someone's post about their job where Agentic AI use is tracked for performance reviews, so they’re now dumping every question they can into the LLM just to pad their metrics and burn as many tokens as they can in the process.
All I can say is... I think maybe a lot of people have had the same idea. 😇
#AgenticAI

Dave Thomas is one of the authors of The Pragmatic Programmer, one of the key people who brought Ruby from Japan to the western world (through "the pickaxe book", Programming Ruby),  one of the authors of the Agile manifesto, and one of my software heros, among other things.
Like many of the original Agile manifesto writers, he's pretty excited about how LLMs/Agents are opening up more possibilities, not ending the software profession.
"Castles in the Air: It's Still Just as Rewarding
It's Still Programming
I was expecting to hate using Claude. I just knew it would dehumanize the process, draining away all the fun stuff, turning what was once creative into a mechanical slog.
I was wrong. Coding with AI is fun. In fact, to me it feels like I'm having more fun than I have had in a while. So I took some notes over the last month to work out why.”
#agenticAI
https://media.pragprog.com/newsletters/2026-03-25.html

Der Guardian berichtet: KI ignoriert bereits Anweisungen. Agentic AI setzt nur noch eins drauf und handelt gleich selbst. Upgrade vom „Ich weiß es besser“ zum „Ich mach das schon“. Für uns bleibt die Rolle des Zuschauers.
#KI #AgenticAI

This seems pretty accurate #agenticAI

Lieschen Müller und Agentic AI mit Open Claw und Moltbook
"Da draussen sind tausende Menschen, die ohne Sicherheitsverstaendnis Systemzugriffe an Tools verteilen, Accounts anbinden, API-Keys reinkippen und sich dann wundern, wenn irgendwas schiefgeht. " @… zu #AgenticAI