10+ years in animation, went all in on AI for my own series. Here's the pipeline behind my latest short
An animator with over 10 years of experience transitioned to AI for their YouTube series, "Woolings World." For a recent short, they used Meshy to create environments and "pose packs" for character placement. A custom plugin, developed with Claude, allowed for easy pose switching and rendered shots using a headless version of Maya, streamlining the camera placement process. The animator shared a link to the short: https://youtube.com/shorts/73dOsggQozg?feature=share.
Time & source
Times shown in UTC
Display time zone: UTC
Local time zone unavailable; showing UTC.
PublishedOffset at this time: UTC+0Sep 12, 2026, 05:21 UTC
IngestedOffset at this time: UTC+0Sep 12, 2026, 18:01 UTC
- Published
- Sep 12, 2026, 05:21
- Ingested
- Sep 12, 2026, 18:01
- Source type
- Dev community
- Tier
- Community
- Source status
- Healthy
Tier is a per-source editorial setting, not a per-item score.
Register to choose your default translation language in settings.
I've been in the animation industry for over 10 years and made the leap about a year and a half ago to go all in on AI and start my own series. I've been working on it for a while and it's called Woolings World on YouTube. I've played with so many pipelines, and the one I just used for this little short was pretty awesome.
I always struggled with camera placement and wanted a quick way to lay out my shots. I used Meshy to make my environment, but then rendered out a "pose pack" to also run through Meshy to get a series of little dolls I could place around. I made a plugin with Claude so that I essentially had a little rig with some swatches of poses I could easily switch out for each shot. Claude would then render all my shots for me using a headless version of Maya.
I then set up a reference pack for how the show should look, and Claude ran my renders through my Google API code to essentially work as my offline renderer.
The really cool thing was I set up a Kitsu site, which helped me communicate my notes with Claude. It works basically like ShotGrid, but I just put my comments on my shots and images, then tell Claude my notes are in and it will reprocess. This is not just for images, but also with my ByteDance key to plug right into Seedance for my final renders.
I'm sure this will evolve over time, but I'm super happy so far and I'll keep you posted on my developments!
Edit: Some folks have requested a link, here you are so you don't have to hunt it down. https://youtube.com/shorts/73dOsggQozg?feature=share