@pirroh

Article

Everyone talks about Continual Learning as if it means one thing only: updating model weights. But there's an inconvenient truth about the agent ecosystem — the vast majority of agents in production today leverage closed frontier models. When you don't own the weights, you certainly can't fine-tune ...

@0xCodila

Article

<b>Most people use AI the same way they used Google in 2005 </b> - Type something, read what comes back, type again The AI sits there doing nothing until you push it - <b>you are the engine</b> - The AI is a wrench you pick up and put down between every turn....

@orcarmi

Thread

- מה היה, ומה יהיה -   מאז אתמול, סוערת הארץ סביב הכרזת הממשלה לפיה היא ״לא מכירה״ בהחלטת בג״ץ בנוגע למועצת הרשות השנייה, אשר לשיטתה נמצאת ״בניגוד חזיתי ללשון חוק מפורשת״ ומהווה ״רמיסה של דבר המחוקק״.   בואו ננסה להבין על מה לכל הרוחות מדובר, וגם להעריך מה צפוי לנו בעתיד. &gt;&gt; ...

@h100envy

Article

A concrete loop engineering case. Not "tune RAG by hand," but build a loop that searches configurations itself, measures recall on an eval, and stops when it hits the goal. With full code....

@craig_certo

Article

I spent some time relaxing yesterday and reading Cameron Wolfe’s @cwolferesearch writing on <a target="_blank" href="https://cameronrwolfe.substack.com/p/agentic-rl" color="blue">Agentic RL: Frameworks and Best Practices</a>. I actually printed it out to read it by the pool and disconnect from devic...