Alexing

Making AI part of everyday life | Lawyer & Dad of two | An ordinary person's take on AI, with the internet's hottest AI stories explained in plain words

[Alexing @ufoyingle]

Over the past couple of years, I've become convinced of one thing: AI isn't just a tech-world thing anymore. It's walking into ordinary homes. Kids' homework, grown-ups' jobs, small-town shops: it's all changing.

So I started my AI Notebook. Here's what I do:
① Every day, I pick the hottest AI stories from around the world and explain, in plain words, why they matter to people like us
② I share my own real experience using AI, fails included

Curious about AI but not a tech person? Follow along. We'll figure it out together.

[Alexing @ufoyingle]

damn

[Niko Veit @coldniko]

Deepseek-V4.1-Flash now runs on 64GB of RAM with Strata.

Using GSQ-RCO quantizations by @mhelcig

[Alexing @ufoyingle]

@longkinght @servasyy_ai 不是我不理解超越在线 API，你是不是没用过 DeepSeek-V4.1 flash 的 API 啊？两百、三百甚至四百的速度都是稀松平常的。不是说 70 多不能用，但和官方在线比起来肯定不是一个量级

[Alexing @ufoyingle]

@xioxing93744764 @Kiana_Liang0609 @mediastorm_en 真诚才是大杀器。何同学也一样都是都是录两个语言版本的

[Alexing @ufoyingle]

@lemomo_ai 顶礼膜拜

[Alexing @ufoyingle]

@Kiana_Liang0609 @mediastorm_en Because they filmed two videos—one in Chinese and the other in English.

[Alexing @ufoyingle]

wow

[Josh @jjpcodes]

This is the most useful prompt I've found in months. I've been using it with Opus 5.5 as a skill called /better to great effect. Grab it here: https://t.co/owQGz21OR7

Here's how it's really stretches the model's performance vs usual prompts:
1. The "what a top professional in the field would do" somehow reminds the model to think of a lot of stuff it wouldn't otherwise do, and (unless this is your core area of expertise), you wouldn't know either.
Obviously Claude/Codex always knows the best practices already, but often it needs reminding to use them (despite it being in the training data), so this helps.

2. Rawest material - models love using proxies for stuff (e.g. Claude sampling logs instead of reading them); so telling them to use rawest materials helps them go back to primary sources/think from first principles, avoiding a lot of problems that come from lazy models and the resulting indirection.

4. Designing independent success criteria: important for the model to know how to actually work; and forcing it to define them means you get better and stronger results vs just letting it vibe stuff. The risk here is that you end up with "proof porn" - think endless hashing or comparisons without testing the real result - all frontier models as of Oct 2026 (Astra 6/Sol 6.1/Opus 5.5/Fable 5.1) all love doing this, so keep an eye on them.

[Alexing @ufoyingle]

@fxtrader 极其傻逼，真的看多了真的想骂人。我原先还以为是我的 GPT 被降智了，那现在看起来大家���一样嘛。这种毫无营养的废话，真的不理解怎么会训练成这个鸟样。

[Alexing @ufoyingle]

@PMbackttfuture opus 是叫 "哦 price"吗？怪怪的

[Alexing @ufoyingle]

@sharenow_today looks nice

[Alexing @ufoyingle]

"I might be one of them," says the guy building one.

As a dad, this is the AI risk I actually think about. Not robots. My kids asking a chatbot for advice and finding it more convincing than me.

One habit I want them to learn: if AI talks you into something big, check with a real person first.

[Elon Musk @elonmusk]

So many people are going to get one-shotted by super persuasive super intelligences and I might be one of them

[Alexing @ufoyingle]

My AI said "no one talks in this footage."

I put on headphones. My kid and I were chatting in almost every clip. It was the best part of the trip. https://t.co/VUPpcqiZMV

[Alexing @ufoyingle]

Claude now writes about 90% of the code behind Claude Code, according to Anthropic's own dev team.

And quality didn't drop. From June to September, merged code changes more than doubled, while the share that had to be rolled back fell by over 40%.

My non-programmer takeaway: the human job is shifting from writing code to checking it.

That's the skill I want my kids to learn. Not just how to make things, but how to tell whether what AI made is actually right.

[ClaudeDevs @ClaudeDevs]

Claude writes about 90% of Claude Code's code.

From June to September, merged PRs more than doubled while the share reverted fell by more than 40%.

We wrote about how we changed code review to get there, and how to adopt it:

https://t.co/9sZe64g3mb

[Alexing @ufoyingle]

Every family has that one person who fixes the TV for mom and dad. This guy handed the job to AI.

His parents' TV showed a sponsored ad before you could even get to Netflix. So he gave Claude Code the model number. It found the TV on the Wi-Fi, removed 25 Xiaomi packages, and wrote a brand-new home screen.

What gets me: this isn't really coding anymore. It's a household chore.

One caution if you're tempted to try this on your own parents' TV: letting an AI delete system apps can break things. Make sure you know what it's removing.

[Gathin (Ø,G) @gathint]

Vibe coded my parents' TV this afternoon!

It was showing a sponsored ad before you could even get to Netflix.

Gave Claude Code the model number, it found the TV on wifi, removed 25 Xiaomi packages and wrote a new launcher.

Looks like an Apple TV now lol

[Alexing @ufoyingle]

@lemomo_ai 感谢告知

[Alexing @ufoyingle]

@realWeZZard 现在纠结的是就 256g 还是换 512g，还是说两台 256g。��要目前技术路线还看不出来，万一两台 256g 不仅可以 double 内存池，还能算���加倍呢

[Alexing @ufoyingle]

@realWeZZard 你们怎么都能正常使用 claude，到底怎么设置使用环境的啊

[Alexing @ufoyingle]

@realWeZZard 我真心觉得 sol6.1 太弱智了

[Alexing @ufoyingle]

@bot @bot 来领邮箱