Hacker Newsnew | past | comments | ask | show | jobs | submit | netless's commentslogin

recent NVDA events are actually bullish for ANT, as GPUs wont be as easily available to chinese proxy data centers in southeast Asia as they were.

2nd bullish thing seems to be that chinese labs are having difficulty in distilling latest frontier models. No deepseek pro for some time now!?

also government, defence, financial and similar institutions and companies will almost sure never use chinese open weights models.


similarly, other nations including the US's staunch allies will never use closed models for government and defence, so that doesn't mean anything.

using claude code with pro subscription and deepseek api. claude is project manager and deepseek is developer. combo seems pretty effective and efficient.


very simple; claude code on pro subscription acting as a project manager, deepseek acting as developer


And not to forget Amiga E language...really cool stuff


DeepSeek Harness works very well, also used Codewhale which is not too shabby


been mostly using WhatsApp, should i be worried?


> should i be worried?

The question implies you already are. But yes, obviously.


had blue N9, people would randomly ask me what model is that?

https://www.domusweb.it/en/design/2011/12/15/portable-cathed...


https://stockchat.sh - a local stock research app you ask in plain English; the numbers come from code over SEC filings and daily prices on your machine, the model only picks which tool to run.


please do not use the "bloomberg for" thing as a marketing phase. You fundamentally misunderstand what the core value of the bloomberg terminal is. This is not it. Its a basic LLM tacked on to financial data. That doesn't make it bloomberg for anything.


ok, i stand corrected.


agree 100%


Surreal


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: