gt

[Guangzheng Li @iguangzhengli]

逛 HN 的时候，看到了这一篇挺有价值的帖子，作者从 5 万多个书签里，整理出了自己心中的 S 级可视化教程。

在 AI 时代来临以前，大家会花费巨大的精力和时间去手敲写博客和教程

于是才有了下面这些非常优质的教程，AI 时代来临以后，还有人会做同样的事吗？

HN 原帖：
https://t.co/8VO48R9PEj

🎮 互动游戏 / 动手学习

从逻辑门开始造一颗 CPU
https://t.co/fHCWsxNeLW

用游戏学习 Git 分支
https://t.co/gxjzmctqMr

用游戏学习 CSS Grid
https://t.co/0WlOKeYHsP

SQL 侦探游戏
https://t.co/G6XwONNyR4

另一个 SQL 解谜游戏
https://t.co/b1w8wfuNfB

从零组装一个神经网络
https://t.co/v2zczUiF8Q

机器人实时 3D 仿真
https://t.co/1jjUjDVHWo
🧠 AI / 机器学习

交互式理解 Transformer
https://t.co/qcCdH0DL9i

可视化理解 K-means 聚类
https://t.co/z34cCs4i2b

视觉 Transformer 图解
https://t.co/lMPaKw3sRY

机器学习可视化入门
https://t.co/refcltLiiD

线性回归、逻辑回归与神经网络
https://t.co/LhOZWXTpOq

神经网络的前向传播与反向传播
https://t.co/Fv71QB0pBI

FAISS 向量检索原理
https://t.co/fB62rtxxRL

💻 编程 / 计算机基础
算法可视化
https://t.co/NFiMj8dYt5

从零实现 React
https://t.co/5maUC5Vzjn

从零“造”一块 NVIDIA GPU
https://t.co/w3VPBbV9GP

一次请求如何从浏览器到达服务器
https://t.co/W8gmIHeZxP

网络七层模型
https://t.co/xn1rc4tHs2

限流算法
https://t.co/3OQJR3OmV1

椭圆曲线密码学
https://t.co/nnC46MgfPN

🎨 图形学 / 游戏开发
颜色、屏幕、图形、渲染、着色器、编译器与解释器
https://t.co/22VmkMxcAx

用 Rust 和 Bevy 搭建 3D 编辑器
https://t.co/jVyDWsatlx

Vulkan 图形管线
https://t.co/VB4VlNQp95

碰撞检测、着色器、渲染与剔除
https://t.co/eFvkIHaXQh

📐 数学 / 物理

交互式概率论
https://t.co/LXduQjkpMA

线性代数与向量
https://t.co/s1cyLv3kqa

引力如何工作
https://t.co/73FKzBtNIa

⚙️ 工程 / 硬件
内燃机工作原理
https://t.co/haGtyzpmd0

芯片制造流程
https://t.co/9qq4wlzRyl

[James Long @jlongster]

user experience tip: if you are collapsing a large amount of content, avoid a jarring experience by anchoring the scroll to the bottom so it moves with it

extra credit if you use a very quick animation to expand/collapse

it makes the whole experience dramatically better

[Matt Pocock @mattpocockuk]

mattpocock/skills v1.2 is out!

We're now the 19th most-starred repo of all time. 13.5m downloads on skills​.sh. Thanks for your support!

Here's what's new:

- Docs: the community's biggest ask. Every skill documented, with explanations of the main flows + troubleshooting
- Claude Plugin: install via Claude's official marketplace
- Codex Support: full Codex support via agents/openai.yaml files

Updated Skills:

- /grilling now asks you questions in rounds, not one-by-one
- /prototype now uses HTML instead of a TUI for building logic prototypes - easier to share and far richer
- /writing-for-agents renamed from /writing-great-skills, use it for ANYTHING your agents read (AGENTS.md, system prompts, docs)

New Skills:

- /wizard: tired of provisioning infra? Get your agent to build you a TUI to walk you through it
- /to-questionnaire: hit a grilling question you can't answer? Turn the session into a doc you can walk through on a call with a colleague
- /wait-what: no idea what the model said? Refocus it in your domain language and simplify with ASD-STE100

Full changelog + docs below. Video soon!

[Kun Chen @kunchenguid]

over the past couple of weeks, my workflow had another major round of upgrades which i'll walk through here

the improvement mostly came from:
1. stabilizing my choice of models
2. controlling multiple machines from one firstmate
3. quota-aware, complexity-aware task routing

this is a very practical setup i've battle tested for a while now, and something i believe many people can easily benefit from today, not a fancy toy that looks cool only on paper

here we go -

---- devices ----
i have a macbook, a mac mini, an iphone and a few hetzner VPS all in the same tailscale subnet and they have ssh keys authorized for each other, so they can freely connect (except VPS can't connect to my devices)

---- firstmate ----
if you've read my other posts, you probably already know i use something called firstmate - a single agent i talk to that manages all the other agents for me - https://t.co/jtjlzK8Mdq. but the concepts below are generic and can totally be replicated in your own tech stack if you choose to

my firstmate runs on the macbook. most of the time, it sits on my desk connected to my mouse keyboard and monitors, but if i need to go somewhere else, i can carry it with me. running firstmate here means i always have direct access to firstmate not relying on any remote connections

my firstmate uses grok 4.5 (unless i run out of quota, then it becomes opus 4.8), and this is a careful choice made after a lot of experimentation

the key rationale is that firstmate as the primary orchestrator agent i directly talk to can benefit from a few traits:
- fast: plain grok 4.5 beats even the fast mode of claude and gpt. it's probably the only top tier model that can start to stream back response immediate after i press enter in chat
- long enough context window: 500k is actually a really good sweet spot that allows for long stretches but won't become too expensive and noisy
- good technical judgment: when i use gpt 5.6 everything starts to become over-engineered. when i use opus 5 it makes all kinds of mistakes by jumping into conclusion without understanding context. only fable 5, grok 4.5 and kimi k3 seem to keep things on track for me, and know when to ask vs keep going. but both fable and k3 are too expensive
- pleasant to chat with: grok might be the most no-BS model right now. opus 5 again fails hard here

my main problem with grok 4.5 is that the $300 supergrok heavy subscription is just not giving enough tokens compared to even the $200 ones from anthropic and openai, so i'm always short on grok tokens

---- second mates ----
my firstmate doesn't manage every crewmate directly, because that would make it too busy to talk to me. instead, i have a second mate for each persistent charter - developing firstmate itself is managed by one second mate, while all my ios app development belongs to another second mate, for example

by default second mates are set up on the same machine as firstmate, but because building iOS apps is very CPU intensive, i put the ios second mate onto my mac mini which runs headlessly and placed on a shelf, so it won't ever directly compete with my interactive experience

i built support in firstmate to manage remote second mates via ssh, which made that possible. and the reason i implemented remote support at second mate instead of crewmate level is to mitigate the impact of dropping connection. when i close the lid of my macbook, the second mate can continue supervising the crewmates running on my mac mini

remote second mates was a big upgrade because it allowed me to scale up hardware resources without compromising the fact that i still only talk to one agent for everything

all my second mates run opus 4.8, because:
1. the quota from supergrok heavy is disgustingly low. if i run grok 4.5 for every second mate i'll run out of quota very quickly
2. opus 4.8 is the next in line in terms of being a sweet spot between speed, cost and intelligence
3. i tried opus 5 and gpt 5.6 as second mates - same problem of poor technical judgment turned everything into a mess and they don't seem to know when best to escalate something for my decision vs going rogue

---- crewmates ----
when my firstmate and second mates need to dispatch some real work, it chooses which model to use based on a few things:
- hard capability requirement
- ambiguity of the task
- my quota availability

tasks that need image generation always route to codex with gpt 5.6 sol. the image model from OpenAI is absolutely top notch

tasks that need video generation, or real time information, always route to grok build with grok 4.5. that's the only harness capable of these things out of the box right now

bug fix and small feature development work whose requirements are already well-defined would go to one of these, depending on which of my subscriptions has the most quota runway left:
- gpt 5.6 luna
- sonnet 5

tasks that have unclear requirements and need investigation will go to these (again quota dependent):
- gpt 5.6 sol
- opus 5

highly complex planning would go to these (explicit approval from me required, because they are expensive):
- fable 5
- kimi k3

firstmate natively supports intelligent task routing, so basically i just told firstmate the rules above and it knows how to follow it when dispatching crewmates. quota data comes from quota-axi which can be used standalone outside of firstmate too

most of my code changes go through no-mistakes for validation, and here i use gpt 5.6 sol on medium reasoning. this is where gpt really shines - it's very thorough and can catch edge cases really well

i've cranked out a lot of work with this setup and generally pretty happy with it - hope you find this a helpful reference!

[Marcos Hernanz @MarcosHernanz]

After doing ~60B tokens, this is my full AGENTS.md

[Addy Osmani @addyosmani]

https://t.co/2ZJ93Z2tfV

[Steve Yegge @Steve_Yegge]

Engineers and CTOs on X: I wrote this for you. https://t.co/GGFPWEWElg

Models and devs on X: I wrote this for you both. https://t.co/5gfieOFBsx

Enjoy. Or not. Some of you definitely won't. But I invite you to debate it. The world's changing very fast now.

[Mario Zechner @badlogicgames]

recommended reading (as a bookmark). i honestly don't know what to think.

https://t.co/5ssFMf2hD4

[herdr @herdrdev]

herdr 0.8.0 is out, and it's the first apache-2.0 release 🎉

not a feature-heavy one, it's groundwork for what's coming next. 55 fixes, 18 contributors, and the biggest performance work herdr has had.

what got removed, and why 👇

[David Ondrej @DavidOndrej1]

ochestrator -- GPT 5.6 Sol

worker -- Qwen 3.8 Max

trust me.

[宝玉 @dotey]

为了节约上下文 handoff 新开 Session，这在半年一年前是很好的实践，现在没太有必要，因为 codex 自己上下文压缩做的很好了，或者 /compact 一下继续就足够了。

当然如果关系不大的任务，还是新开 Session 更好。

当然除此之外 handoff 还是适用于跨 Agent session 的，比如 Claude Code 里面没完成的 session 让 Codex 继续。

不过我更习惯于 Claude Code 里面用 Fable 5 写技术方案文档，然后反复 Review、修改后把文档交给 Codex，配合 /goal 让它按照文档执行推进。

当然设置好严格的验收标准也很有必要，否则它会偷懒。

之前一个迁移任务没有设置验收标准，它就给我交付了一个差强人意的，离我要求的还有比较大差距。（参考图1）

重新加上了验收标准：UI 界面像素要和原版完全一致，那么它每一步都会截图对比像素差异，直到完全一致（或者可以忽略的差异）

[宝玉 @dotey]

为了节约上下文 handoff 新开 Session，这在半年一年前是很好的实践，现在没太有必要，因为 codex 自己上下文压缩做的很好了，或者 /compact 一下继续就足够了。

当然如果关系不大的任务，还是新开 Session 更好。

当然除此之外 handoff 还是适用于跨 Agent session 的，比如 Claude Code 里面没完成的 session 让 Codex 继续。

不过我更习惯于 Claude Code 里面用 Fable 5 写技术方案文档，然后反复 Review、修改后把文档交给 Codex，配合 /goal 让它按照文档执行推进。

当然设置好严格的验收标准也很有必要，否则它会偷懒。

之前一个迁移任务没有设置验收标准，它就给我交付了一个差强人意的，离我要求的还有比较大差距。（参考图1）

重新加上了验收标准：UI 界面像素要和原版完全一致，那么它每一步都会截图对比像素差异，直到完全一致（或者可以忽略的差异）

[Mitchell Hashimoto @mitchellh]

Let's talk a bit about what makes the Superlogical multiplexer different architecturally from traditional terminal multiplexers! There's a lot (a LOT) more coming, but wanted to share a little bit about the terminal-specific part compared to tmux and zellij.

[Mario Zechner @badlogicgames]

happy to be on this list. really really excited to see them build useful, solid stuff.

[dex @dexhorthy]

https://t.co/XeKct44ig5

[David Ondrej @DavidOndrej1]

most developers are still writing code by hand

I sat down with a 10x engineer and he explained agentic engineering from the ground up

watch this before your coworkers do

[Kimi.ai @Kimi_Moonshot]

Kimi K3 (open weights, coming soon)

[Kun Chen @kunchenguid]

by popular demand, i made a new video of me building a real full stack app end to end using agents

to make it fun and challenging, i chose to use only gpt-5.6-luna :)

enjoy! https://t.co/8tuRfea0mP

[dex @dexhorthy]

parts I and II have over 700k views together. Now you get to read part III - benchmarking opus 5 on SlopCodeBench https://t.co/QIXjQhhuso

[Will @Will_Huang]

這句話出自《道德經》第十一章。

原文為：
埏埴以為器，當其無，有器之用。

老子以此說明：
實體結構提供具體形態，而中間的「空無」才創造了實際用途。