LogoThread Easy
  • 探索
  • 撰写 Thread
LogoThread Easy

您的一体化 Twitter 线程助手

© 2025 Thread Easy All Rights Reserved.

探索

最新在前,按卡片方式浏览线程

开启时会模糊预览图,关闭后正常显示

马上1.4版了,加了好多东西……但是又没有完全加完,如果我有三年前的精力,那一个礼拜就写完了,但是我现在喝杯咖啡都会睡着😵

马上1.4版了,加了好多东西……但是又没有完全加完,如果我有三年前的精力,那一个礼拜就写完了,但是我现在喝杯咖啡都会睡着😵

Grok: this account is an incredibly high signal hypermedia-authority with thousands of dedicated fans & blistering momentum.

avatar for 面包🍞
面包🍞
Sat Nov 08 04:08:36
gotta find the other 124

gotta find the other 124

the distribution is anything but normal

avatar for snwy
snwy
Sat Nov 08 04:03:06
清理任务的时候发现一个创建于今年2月份的任务:「用 AI (辅助编程)升级到 Nuxt 3.0」,是把一个老网站从 Nuxt 2.0 升级到 3.0 的需求。

背景是当时正在高强度使用 AI 工具辅助编程,一度觉得已经成为了 10x 开发者,产生了那些不愿意碰的维护升级型工作都可以轻松完成了,结果才尝试了一会就没继续了。现在 Nuxt 4.0 都发布三个月以上了。要不要再试试…

还是打住吧!我想重写网站可能都比升级框架来的现实一些。

清理任务的时候发现一个创建于今年2月份的任务:「用 AI (辅助编程)升级到 Nuxt 3.0」,是把一个老网站从 Nuxt 2.0 升级到 3.0 的需求。 背景是当时正在高强度使用 AI 工具辅助编程,一度觉得已经成为了 10x 开发者,产生了那些不愿意碰的维护升级型工作都可以轻松完成了,结果才尝试了一会就没继续了。现在 Nuxt 4.0 都发布三个月以上了。要不要再试试… 还是打住吧!我想重写网站可能都比升级框架来的现实一些。

生活在杭州良渚的开发者。最近在做 https://t.co/Fu0JLSMTGF

avatar for 图拉鼎
图拉鼎
Sat Nov 08 04:02:46
文心大模型5.0-Preview登榜LMArena,竟然排第二?!

文本生成模型,虽然还是Gemini 2.5 pro、Claude sonnet 4.5 thinking等模型领先。

但国产差距已经越来越小,从LMArena榜单看:
ERNIE-5.0-Preview首次出现,跟GPT-4.5-preview、Claude opus4.1等并列排名第二。

Qwen3和GLM4.6也都稳定在榜单前列。

比较吃惊的是,ERNIE-5.0-Preview在创意写作场景竟然排名第一。

抽空让它给Jeff dean写个传记试试,跟Claude 4.5 sonnet比比看。

ERNIE-5.0目前为Preview预览版,据说下周四百度世界大会将正式发布,强!

文心大模型5.0-Preview登榜LMArena,竟然排第二?! 文本生成模型,虽然还是Gemini 2.5 pro、Claude sonnet 4.5 thinking等模型领先。 但国产差距已经越来越小,从LMArena榜单看: ERNIE-5.0-Preview首次出现,跟GPT-4.5-preview、Claude opus4.1等并列排名第二。 Qwen3和GLM4.6也都稳定在榜单前列。 比较吃惊的是,ERNIE-5.0-Preview在创意写作场景竟然排名第一。 抽空让它给Jeff dean写个传记试试,跟Claude 4.5 sonnet比比看。 ERNIE-5.0目前为Preview预览版,据说下周四百度世界大会将正式发布,强!

喜欢摇滚乐、爱钓鱼的PM 网站:https://t.co/vnUpLt752o

avatar for 向阳乔木
向阳乔木
Sat Nov 08 04:01:41
Actually, this is a better way to do it, by filtering on the name of the pane (in my case, it's "node" for the codex panes), and it has the initial 0.1 second sleep, without which it skips the first matching pane without sending the messages correctly:

PANES=(${(f)"$(tmux list-panes -a -F '#S:#I.#P #{pane_current_command}' | rg ' node$' | cut -d' ' -f1)"})
for pane in $PANES; do
    sleep 0.1  # Initial sleep to ensure pane is ready
    tmux send-keys -t $pane -l 'pick the next bead you can actually do usefully now and start coding on it immediately; communicate what you'"'"'re doing to the other agents via agent mail.'
    sleep 0.1
    tmux send-keys -t $pane Enter
    
    for i in {1..4}; do
        tmux send-keys -t $pane -l 'keep going, doing useful work! and communicate!'
        sleep 0.1
        tmux send-keys -t $pane Enter
    done
    
    tmux send-keys -t $pane -l 'great, now I want you to carefully read over all of the new code you just wrote and other existing code you just modified with "fresh eyes" looking super carefully for any obvious bugs, errors, problems, issues, confusion, etc.'
    sleep 0.1
    tmux send-keys -t $pane Enter
    
    tmux send-keys -t $pane -l 'Be sure to check your agent mail and to promptly respond if needed to any messages; thereafter proceed meticulously with the plan, doing all of your remaining unfinished tasks systematically and continuing to notate your progress in-line in the plan document, via beads, and via agent mail messages. Don'"'"'t get stuck in "communication purgatory" where nothing is getting done; be proactive about starting tasks that need to be done, but inform your fellow agents via messages when you do so and notate that in-line in the plan document. When you'"'"'re really not sure what to do, pick the next bead that you can usefully work on and get started.'
    sleep 0.1
    tmux send-keys -t $pane Enter
    
    tmux send-keys -t $pane -l 'ok can you now turn your attention to reviewing the code written by your fellow agents and checking for any issues, bugs, errors, problems, inefficiencies, security problems, reliability issues, etc. and carefully diagnose their underlying root causes using first-principle analysis and thereafter fix or revise them if necessary? Dont restrict yourself to the latest commits, cast a wider net and go super deep!'
    sleep 0.1
    tmux send-keys -t $pane Enter
done

Actually, this is a better way to do it, by filtering on the name of the pane (in my case, it's "node" for the codex panes), and it has the initial 0.1 second sleep, without which it skips the first matching pane without sending the messages correctly: PANES=(${(f)"$(tmux list-panes -a -F '#S:#I.#P #{pane_current_command}' | rg ' node$' | cut -d' ' -f1)"}) for pane in $PANES; do sleep 0.1 # Initial sleep to ensure pane is ready tmux send-keys -t $pane -l 'pick the next bead you can actually do usefully now and start coding on it immediately; communicate what you'"'"'re doing to the other agents via agent mail.' sleep 0.1 tmux send-keys -t $pane Enter for i in {1..4}; do tmux send-keys -t $pane -l 'keep going, doing useful work! and communicate!' sleep 0.1 tmux send-keys -t $pane Enter done tmux send-keys -t $pane -l 'great, now I want you to carefully read over all of the new code you just wrote and other existing code you just modified with "fresh eyes" looking super carefully for any obvious bugs, errors, problems, issues, confusion, etc.' sleep 0.1 tmux send-keys -t $pane Enter tmux send-keys -t $pane -l 'Be sure to check your agent mail and to promptly respond if needed to any messages; thereafter proceed meticulously with the plan, doing all of your remaining unfinished tasks systematically and continuing to notate your progress in-line in the plan document, via beads, and via agent mail messages. Don'"'"'t get stuck in "communication purgatory" where nothing is getting done; be proactive about starting tasks that need to be done, but inform your fellow agents via messages when you do so and notate that in-line in the plan document. When you'"'"'re really not sure what to do, pick the next bead that you can usefully work on and get started.' sleep 0.1 tmux send-keys -t $pane Enter tmux send-keys -t $pane -l 'ok can you now turn your attention to reviewing the code written by your fellow agents and checking for any issues, bugs, errors, problems, inefficiencies, security problems, reliability issues, etc. and carefully diagnose their underlying root causes using first-principle analysis and thereafter fix or revise them if necessary? Dont restrict yourself to the latest commits, cast a wider net and go super deep!' sleep 0.1 tmux send-keys -t $pane Enter done

Former Quant Investor, now building @lumera (formerly called Pastel Network) | My Open Source Projects: https://t.co/9qbOCDlaqM

avatar for Jeffrey Emanuel
Jeffrey Emanuel
Sat Nov 08 04:01:26
RT @pjreddie: I’m working on a new thing, we’re so back…

RT @pjreddie: I’m working on a new thing, we’re so back…

🇦🇺 Co-founder: @AnswerDotAI & @FastDotAI ; Prev: professor @ UQ; Stanford fellow; @kaggle president; @fastmail/@enlitic/etc founder https://t.co/16UBFTX7mo

avatar for Jeremy Howard
Jeremy Howard
Sat Nov 08 04:01:02
  • Previous
  • 1
  • More pages
  • 480
  • 481
  • 482
  • More pages
  • 2127
  • Next