AI coding tools are everywhere, but most still function as glorified autocompletes, great for snippets and boilerplate, but when it comes to building a fully functional app, they fall short.
You still have to manually stitch together frontend, backend, database, and deployment, often fixing AI-generated code along the way.
So we asked: What if AI could do more than just generate code? What if it could handle the entire dev workflow, end to end?
For the past 1.5 years, we’ve been testing this idea with 20,000+ developers, working through the real challenges of AI-powered development:
> Understanding an entire project instead of just the last few lines of code.
> Handling real-world app logic, not just generic "hello world" templates.
> Connecting everything seamlessly, from database models to APIs to the frontend.
> Deploying instantly, without the usual debugging loop that makes AI-generated code frustrating.
What if AI could actually build apps — not just generate code snippets, but scaffold an entire working product from scratch?
So we built saas-builder, an open-source CLI that lets you describe your app idea and generates a full-stack scaffold using Next.js and Supabase.
You run: saas-builder init …and five AI agents take over:
> UI Agent → builds a responsive frontend > Auth Agent → wires up Supabase Auth > DB Agent → creates schema, migrations, and types > Terminal Agent → installs dependencies & folders > Browser Agent → launches the app locally
What you get is a clean, typed, ready-to-deploy app.
It’s still early days — very much a side project. But it’s already saved us hours of boilerplate setup.
More importantly, it gave us a glimpse of what agentic development might actually feel like.
We’re releasing it under MIT, installable via pip, and would genuinely love your thoughts:
GitHub: https://github.com/jatingarg619/saas-builder
Appreciate you reading. We’re around to answer questions, chat about ideas or hear whatever feedback you’ve got.