What are they doing now? crisis of ideas and just hyping? pity.
271 karma · joined January 16, 2025
My blog available at https://zhisme.com/
What are they doing now? crisis of ideas and just hyping? pity.
this actually very useful where you can rely your classical programming logic with coefficients provided by jev decision model results if (jevResponseJudge < 0.3) doSomething() else doNothing()
how would u code with LLM in such cases? parse responses for keywords? xd
no, they provide very useful interface. Just try it before writing posts about "hype".
this one has more stars and more popular
[^1]: https://github.com/mame/ai-coding-lang-bench?tab=readme-ov-f...
Yet arguable.
I don’t want to downplay the work done by the maintainers on the contrary, huge thanks to them. But I do feel the version number is a bit misleading.
That said, the work on the ZJIT[1] compiler is massive. It’s serious, professional engineering, and definitely deserves respect.
Did you try other IDEs? For example from Jetbrains? I'm not using IDEs but actually don't know others besides VSCode/Cursor and Jetbrains
I also noticed that copilot nowadays is forcing you to upgrade to their with following text:
"You've reached your monthly code completion limit. Upgrade your plan to Copilot Pro (30-day Free Trial) or wait until 2025-12-19 for your limit to reset to continue coding with GitHub Copilot"
Was using it actually like smarter auto-completion. But paying for that, hell no.
Now every system design interview expects you to build some monstrous stack with layers of caching and databases for a hypothetical 1M DAU (daily active users) app.
Mess in the head.