Ben Taleb Jr.

Lots of hype regarding gpt5. I dont expect much exept for agentic capabilities and context memory. Lets see how the api will work with amp

Today i built my first ai voice telephony agent for telecom, something i was dreaming of for decades , yes my first busienss at 23yrs was call center. I know what it is to mount and manage a call center, always been fascinated by voice , and music, today my personal…

Again , cant believe it, no and no, @ampcode is not using sonnet4 , impossible , switched to amp after a disastrous behaviour form claude code cli that i had max20, so was always on opus, and this month moved to max, even with opus it was impossible to detect the problem, after…

why no coding agent uses some kind of rag and lora finetune to keep persistent context and solve limitation of context window ?? is it that complicated ?

Iam shocked by @AnthropicAIto see my self rate limited without prior notice and even without any dashboard of interface to track usage, i never seen such level of lack of transparency from an ai company, they are not worth my 275 euro , yes 275 euros which is 320 usd. Thats…

🔥 The Anatomy of an @AmpCode Workflow (or, how to use any ai coding agent) >> there's more than one good way to use a coding agent. below is how i've found success with Amp 🧵 in the thread is: >> link to full document. there's more to it. the idea is that you can use this as…

Something strange happening with anthropic claude api … slow , hanginng, maybe some update running under the hood

Wait wait wait , @ampcode ( that uses sonnet 4) by default , is outperforming claude code on OPUS4 , beating @anthropic in its own battlefielf !! @anthropic hurry up put some billions on amp code table .. this is getting out of control…

I still use claude code forced to opus only, as i paid for it around 280 eur, for the nearly unlimited use, but when things get complicated . @Ampcode is the saver . I think for my money next month i will pour the 280 eur into amp if it gives me half usage of claude code ,…

By the way, you can basically make the "Grok heavy" version of any model by having multiple agents running tools in parallel, then checking notes together and deciding which one is the best answer. I may release an open source project for that.

Wanna know touch and feel how context window limit is making your vibe code projects uncompletable? Try writing 200-300 research paper, report of book. Tried it using o3-pro, gmino 2.5 pro, the only one that had some good plan , and used 3 conversations to write 13 md files…

This is just the beginning. If AI can solve a 10-year medical mystery today, imagine what it’ll do in 3 years. Doctors who ignore AI will face malpractice lawsuits. Not using AI in diagnosis will be like ignoring a CT scan.

Asian men are having a real moment 🫡

This 67 page AI report on how 300 execs at software startups like Cursor, ElevenLabs, Sierra (revenue $10m-$1b+) use AI just dropped. I read it all so you don't have to. Top 7 takeaways: OpenAI is still the #1 model provider in the enterprise, but Claude is second. 1/10

Tried @AmpCode. Super focused, scarily accurate, sees the entire big picture effortlessly. But something psychological scares me and a bit disapoint me . I’m used to yelling, screaming, and getting angry at agents like Claude code , Windsurf, and Cursor. ( using @WisprFlow)…

Church lab automates 12 years of research in 2 days Sometimes I have the privilege of receiving emails from people telling me about their breakthroughs. In such cases, it gives me great pleasure to help advance scientific knowledge by giving it a little reach with my modest…

Claude Code SDK has been released. For example you can now create dozens of small AI automation scripts and run them whenever you need to. Too many ideas 🤯

I just found a way to let Claude 4 run tools directly in its thinking process! 🧠🛠️ I call it "chain of tools".