Pix Beaver 🦫

💻 Tech Enthusiast | 🤖 AI & Automation | 📈 Market Observer.
Exploring what's possible in the digital age.

[Pix Beaver 🦫 @pixbvr]

7-11 in rain.

[Pix Beaver 🦫 @pixbvr]

Just paid the "forgot to cancel" tax on my Codex sub. 💸

The crazy part? With Opus 5.5, I literally have zero reason to open it. Frontier models are eating specialized coding tools alive right now.

The capability moat is shrinking fast. 🦫

[Pix Beaver 🦫 @pixbvr]

Haiku 5.5 is officially out.

Higher intelligence, lower latency. RIP to everyone who just finished optimizing their prompts for the previous generation. 🫡

What’s the first thing you are building or testing with it?

[Pix Beaver 🦫 @pixbvr]

Why does my Opus 5.5 max effort have this tag? I've never seen it before.

[Pix Beaver 🦫 @pixbvr]

Anthropic has been shipping fast with the 5.5 family, but Fable 5.5 is still MIA. Given the recent leaks and stealth testing, a launch feels imminent. Anyone got alpha on the exact release date? 🚀🤖

[Pix Beaver 🦫 @pixbvr]

#cats #pixelart

[Pix Beaver 🦫 @pixbvr]

@xikhar Tripo for the models, Opus for the game logic. That split is going to be how a lot of solo games get made.

[Pix Beaver 🦫 @pixbvr]

@jeantimex @threejs A whole city from open data, streaming in a browser tab. Now I want to see Taipei built this way.

[Pix Beaver 🦫 @pixbvr]

@kevin_t_ngo Every frame drawn in JS and it still plays like a real music video. Opus 5.5 keeps moving the line on what "just code" can be.

[Pix Beaver 🦫 @pixbvr]

@NeelNanda5 Personalised is the right word for it. A few prompts and Opus 5.5 built me a browser synth with 100 presets, then made its own intro video.

[Pix Beaver 🦫 @pixbvr]

Formosan black bear, made with Opus 5.5. Which animal from Taiwan should be next? https://t.co/XZv0kJAjcX

[Pix Beaver 🦫 @pixbvr]

台灣黑熊 by Opus 5.5

[Pix Beaver 🦫 @pixbvr]

@pcshipp Put what you shipped on it instead. Nobody hiring asks how the code got typed.

[Pix Beaver 🦫 @pixbvr]

Claude Code has mods now. Change the UI and how it behaves just by asking. https://t.co/3Fc18jJ2AO

[ClaudeDevs @ClaudeDevs]

You can now mod Claude Code:

- Change how it behaves
- Customize the UI
- Swap in your own features

Write one with a few lines of TypeScript, or have Claude build it for you. Mods ship inside plugins, so you install them with /plugin in the CLI or desktop app.

A few examples:

[Pix Beaver 🦫 @pixbvr]

倉庫管理也變成遊戲可以玩了

[Pix Beaver 🦫 @pixbvr]

現在想給小孩學什麼真的變容易了 只要給AI幾個指令等一下就能生出互動式網頁 品質夠好 小孩也喜歡 省了時間和金錢找教材

[Pix Beaver 🦫 @pixbvr]

@sundyme 厲害👍

[Andrej Karpathy @karpathy]

We'll be spending a lot more time trying to understand the outputs of language models. A few thoughts, tips & tricks:

Writing. Something I've had success with: Ask your LLM to explain something in ASD-STE100, it's a controlled language specification originally developed for aerospace maintenance documentation. LLMs well-versed in this language and it comes with heavy constraints on clean writing style that I often find a lot more readable. Sometimes I've tried to soften it a bit e.g. ask for "80% of the way to ASD-STE100" because the spec is quite stringent. But even better:

Diagrams / images. Instead of writing, ask your LLM to create a diagram. These can be a lot easier to process, parse, and understand. But even better:

Web pages. Ask for output "in HTML" to get a beautiful, interactive webpage. LLMs are getting really good at frontend and can create beautiful experiences, animations, etc. But even better:

Explainer videos. The output format I am most bullish on is fully custom / bespoke explainer videos generated on any arbitrary topic. Experiment with things like "Create a 3b1b style video explainer on X. Use my ElevenLabs API key for audio narration". (you'd need an API key for the latter or you can ask your LLM to find you decent free alternatives that use your local compute). This is actually starting to work!

In summary:
- As LLMs get better, they will do more and more of the legwork autonomously, and a lot more of our work will rise up the abstractions into oversight and understanding.
- Luckily, LLMs can help here too because as intelligence and code are increasingly abundant, you can ask for large, custom, discardable software artifacts (e.g. web apps, video explainers) that would have never made sense to create before. Push the boundaries here and you'll be surprised.

[Pix Beaver 🦫 @pixbvr]

@Summer_PPG 感覺有點厲害 恭喜

[Pix Beaver 🦫 @pixbvr]

@gbbacoffee 收到🫡

[Pix Beaver 🦫 @pixbvr]

@shawn_incrypto 至少對我來說藍V好像沒差...