Show HN: New way to validate your LLM webapp idea and earn on token margins

I've built a whole new UX and platform where you can quickly make LLM-backed webapps and when people use them, you earn on each AI API call. I've been working on this for 2 years! What do you think?

Here's how it works:

1) You make a Project, which means we run a docker container for you that has python/flask and an optional sqlite database.

2) You provide a project name and description

3) The LLM makes tickets and runs through them to complete your webapp.

4) You get a preview iframe served from your docker, and access to server logs and error messages.

5) When your webapp is ready, you can Publish it to a subdomain on our site. During the publish process you can choose to require users to log in via Code+=AI, which enables you to earn on the token margins used. We charge 2x the token costs of OpenAI - that's where your margin comes in. I'll pay OpenAI the 1x cost, then of the remaining amount you will earn 80% and I'll keep 20%.

The goal: You can validate your simple-to-medium LLM-powered webapp idea much easier than ever before.

Some fun technical details: Behind the scenes, we do code modifications via AST transformations rather than using diffs or a full-file replace. I wrote a blog post with details about how this works: https://codeplusequalsai.com/static/blog/prompting_llms_to_m...

Would love some feedback! What do you think?