Augusdin

Rock 🎸 & Code 👨🏻‍💻

Trying to build some cool shit.
Creator of @phewtab

Recent 3-month creations:

[Augusdin @augusdin]

I spent nearly five months developing this extension, aiming to create a product that truly enhances the browsing experience.

Tomorrow, I will officially launch it on https://t.co/T51ly48SBm, hoping to reach a wider audience.

I also hope for your support.

[Phew AI Tab @phewtab]

Do you often encounter this when browsing?

[Augusdin @augusdin]

@GaryMarcus @peterwildeford Let’s just agree both matter —
the model is the brain,
the tools are the hands,

and neither gets much done alone.

[Augusdin @augusdin]

@davidcoffeeboy @GaryMarcus @geoffreyhinton 🤣

[Augusdin @augusdin]

@thsottiaux Codex reasoning is top-tier, but its UI/UX taste is… honestly bad.
I need multiple iterations to reach what Claude Code gets in one shot.

Why is there such a gap in design intuition?

[Augusdin @augusdin]

@Google @GoogleDeepMind When Veo ?

[Augusdin @augusdin]

@kamath_sutra It’s cool, man. But do you know about ChatGPT’s real-time voice? What’s the difference?

[Augusdin @augusdin]

@kirawontmiss This need be update

[Ilya Sutskever @ilyasut]

One of life’s great pleasures is to work with brilliant and kind colleagues

[Augusdin @augusdin]

@ArjunPatel_AI I hope that if I give Claude a portrait image, it can generate a high-fidelity 3D model in Blender.

Is this achievable with current technology? If so, how could it be implemented?

[Augusdin @augusdin]

@businessbarista Claude

[Augusdin @augusdin]

An AI without real-time learning isn’t intelligent.
It’s a simple truth that many choose to ignore.

[Augusdin @augusdin]

@levelsio I’m based in Beijing! If you visit, I’d love to show you around and connect you with some indie devs here.

[Augusdin @augusdin]

@scaling01 This might still be the same path as GPT-4.

When GPT-4 launched, it was slow. Then came GPT-4-Turbo — a distilled version.

It wasn’t great at first, but OpenAI kept refining it until it replaced GPT-4.

GPT-5.1 seems to follow that path, just smarter with the name.

[Augusdin @augusdin]

Making robots walk gracefully—like a human woman—has always been one of the hardest challenges.

I thought it would take 5 more years.

Yet XPENG from China just did it.

The U.S. should face this failure and reflect, not act like Neanderthals doubting it’s a woman in disguise.

[The Humanoid Hub @TheHumanoidHub]

XPENG's next-gen IRON robot effectively crossed the uncanny valley, leading many to believe it was a human in a suit.

In a follow-up event to prove it was a robot, He Xiaopeng had its leg skin cut open in front of a live audience. The robot then walked off the stage.

[Augusdin @augusdin]

@doganuraldesign How can I turn a real photo of a person into this style? I’d really appreciate some guidance.

[Augusdin @augusdin]

@Dr_Singularity This reminds me of Chaplin’s Modern Times.

[Augusdin @augusdin]

@sport11usa code

[Augusdin @augusdin]

@__theSeyram @OpenAI

[Augusdin @augusdin]

@krishdotdev That’s basically what people in China wake up to every day."

[Augusdin @augusdin]

It's still too AI

[Luma @LumaLabsAI]

This is Ray3. The world’s first reasoning video model, and the first to generate studio-grade HDR. Now with an all-new Draft Mode for rapid iteration in creative workflows, and state of the art physics and consistency. Available now for free in Dream Machine.

[Augusdin @augusdin]

@sama We’re grown—why not both?