Vargai/SDK – JSX for AI video, declarative programming language for Claude Code

5 days ago (varg.ai)

> I don't see the point of JSX, seems to decrease type safety

JSX compiles to typed function calls. The type safety is in the component definitions, not the syntax.

The real question is whether video composition is actually hierarchical enough to benefit from JSX. For simple clips, probably overkill. But once you're layering talking heads, captions, b-roll, and transitions, the tree structure starts making sense.

The hardest part isn't the generation. It's the orchestration. Caching, retry logic, stitching outputs from multiple providers. A declarative layer that handles that automatically is genuinely useful.

looking at the code examples i don't see the point of JSX, seems to decrease type safety and typing completion

If you told me this was satire, I might respond that it's too on-the-nose. I really doubt that Spotify would be happy that their logo is front-and-center here.

This is very cool. And very timely. Recently had a discussion on whether we wanted to build out a new video pipeline through weavy or something more declarative like this. Surprised that there don't seem to be many similar tools. I suppose most folks either use weavy or just chain API calls. JSX is an interesting choice too.

  • I am building flickspeed.ai where you can build a pipeline in plain english. Let me know if you'd like to have a demo

> A 30-second video might take 3-5 min first render, 10 seconds cached.

How could a cached video possibly take 10 seconds? I would expect, at most, that it would cost whatever is necessary to read it from disk.

JSX is a convenient notation for structuring HTML that is created with React and similar frameworks. You are imposing a markup hierarchy on attributes that are not inherently hierarchical while giving examples of deceptive weight loss scam slop.

In the past I would say you should be ashamed of yourself but now I don't bother.

I am impressed, i might give it a try. The prices for generation of images and videos scared me a little bit. I thought they were much cheaper

  • Maybe I misunderstand the project but I feel it'd make sense to support some local inference, i.e using arbitrary ComfyUI workflows?

im a product engineer. i dont like building endless workflows in comfy ui or weavy. i always wanted to do it with agent. This sdk helps.

This turns Claude Code into Sora lol

It's similar to remotion.dev, but focuses on generative video. Uses declarative JSX to orchestrate AI calls, which makes it much more readable!

Weight loss scams have been horrible for ages, but that AI generated weight loss product example is some truly dystopian stuff.

We can build a better future with these tools, how about we build it instead of this garbage.

  • Not sure if you are missing the point or hitting on the actual point.

    That is an example. Given the amount of AI-generated scam content exactly like that (and worse, have you seen the ED ones?) on the internet today, creating something in the same vein seems harmless.

    On the other hand, unleashing the ability to generate massive numbers of variations cheaply, and experimenting to find the optimal mixture of insecurity, fear and avarice to exploit....that's dystopian. But Madison Avenue actually got most of that figured out long ago.