I Tried AI Modeling and Became a Xiaohongshu Creator for a Few Days

↗

When I first tried Astra, I wanted to test its modeling and video-generation capabilities. 3D was unfamiliar to me. I like Doro and those goofy cats, so I used them for the experiment, working with Blender.

The Doro result was fairly ordinary. The goofy cat turned out better: it has that vacant little look. I also tried making Doro’s body. After modeling, I had AI help make videos, too.

Models and animations

Project page (opens in a new tab) · Source (opens in a new tab)

More models are on the experiment page (opens in a new tab).

A few days as a Xiaohongshu creator

After these experiments, I spent a few days being a Xiaohongshu creator and got roughly 3k likes. Watching my phone keep pinging was pretty nice.

Then I realized 3k was still tiny. I’d been living in my own little world.

The way Xiaohongshu distributes posts was interesting too. My impression was that it would first stop around 200 and give a post a small audience. If people liked, shared, or commented, perhaps it would show the post to more people? If the post was boring, maybe it would just stay at 200.

That’s my guess, of course. It might be wrong.

I think programmers encounter AI often in work and learning, so we’re closer to these tools than most people. There’s no need to sell ourselves short. If something interests us, we can make something with AI. It could be a project, or just something silly to post online. That can be fun too.

I see that as learning, even if it looks different from traditional study. For me, using AI more is how I find out what I actually need. It might also turn up a real problem worth solving.