Vercel Rebuilds v0 to Take AI Apps From Prototype to Production

Vercel has rebuilt its v0 “vibe coding” service to focus on a problem many AI app-building tools still struggle with: turning AI-generated code into software that can ship inside real production environments, not just prototypes.

The original v0, launched in 2024, was designed to help developers get past the blank canvas stage by generating early UI scaffolding from prompts. But while the results could look polished, the output was often disposable—teams typically had to rewrite the code to get it production-ready.

Vercel says more than 4 million people have used v0 to build millions of prototypes. However, it lacked key elements needed for production deployment—an issue that’s become familiar across the fast-growing category of AI-assisted “vibe coding” tools. The gap is especially visible in enterprise settings, where requirements go beyond generating code to include working within established design systems, security policies, and approved infrastructure.

Vercel CPO Tom Occhino according to VentureBeat described the situation as “the world’s largest shadow IT problem,” arguing that AI-enabled software creation is already happening across enterprises. In that environment, credentials may get copied into prompts, company data can flow into unmanaged tools, apps may be deployed outside approved infrastructure, and teams can lose an audit trail.

Vercel also points to the limitations of other tools as an example of the broader challenge. Claude Code, for instance, can generate backend logic and scripts, but it does not deploy production UIs within existing company design systems while enforcing security policies.

What’s new in the rebuilt v0

The rebuilt v0 is generally available now. Vercel says the new version is designed to close the production deployment gap by connecting AI-generated work to existing codebases and deployment setups.

• Imports existing GitHub repositories, so work can start from a real project rather than a fresh prototype.
• Automatically pulls environment variables and configurations to better match how production apps are actually run.
• Generates code in a sandbox-based runtime that maps directly to real Vercel deployments.
• Enforces security controls and proper Git workflows, with Vercel positioning it as a way for non-engineers to ship production code without bypassing governance.

Vercel’s pitch is that v0 still keeps code visible and reviewable, aligning AI-assisted development with the oversight enterprises expect.