🚀 Better Auth is Joining Vercel

Better Auth is officially becoming part of the Vercel family.

We are uniting to speed up the development of open-source authentication and to pioneer the security of agentic workflows.

Beka

The Big Announcement

I am thrilled to share that Better Auth is joining Vercel. Since our inception, the Vercel team has been a primary source of inspiration for us, embodying the very principles that drove us to create Better Auth in the first place.

Our Shared Vision

Vercel is fully aligned with our mission to ensure authentication remains:

Open Source
Framework Agnostic
Platform Agnostic

By combining our strengths, we gain the resources necessary to refine the framework that so many developers now depend on. Furthermore, we are looking toward the horizon of authentication: AI Agents. We aim to build a future where agents can operate on a user's behalf using access that is:

$\text{Secure}$
$\text{Scoped}$
$\text{Revocable}$

We plan to integrate these primitives directly across the Vercel product ecosystem. I want to extend my deepest gratitude to our community and everyone who pushed us toward this milestone.

📖 The Origin Story: How Better Auth Began

Three years ago, while developing an open-source web analytics tool, I hit a wall with authentication. At the time, my default choice for Next.js projects was NextAuth.

The Gap in the Market

For simple needs—like a "Sign in with Google" button and a basic user profile—NextAuth was perfect. However, my analytics project required multi-tenancy (Organizations), which included:

Inviting team members
Defining roles and permissions
Implementing strict access control

Feature	Basic Auth (NextAuth)	Advanced Auth (Needed)
Social Login	✅	✅
User Object	✅	✅
Multi-tenancy	❌	✅
Role-Based Access	❌	✅
Team Invitations	❌	✅

I tried to build these features manually, but after weeks of refactoring, I was still unhappy. I even attempted to create next-org (a wrapper for NextAuth to add organization support), but the API felt clunky and unergonomic. I eventually abandoned it.

The Breaking Point

The frustration grew as I encountered more hurdles:

Expo: Adding OAuth via NextAuth felt nearly impossible.
Svelte: Migrating a project from Next.js to Svelte was a nightmare because there was no "NextAuth for Svelte."

I noticed a recurring debate on X (Twitter) regarding whether developers should "roll their own" auth or use a managed service. The hesitation to build usually came down to: ~~"I don't want to spend months building feature X."~~

I decided to build a framework-agnostic solution: a simple core for basic apps, extensible via plugins for complex needs. I named it better-auth. I told myself that if the npm package name was available, it was a sign to proceed.

The Build and Launch

For seven months, I worked sporadically on the project, creating supporting libraries along the way:

better-fetch: Essential for the Better Auth client.
better-call: A lightweight web server framework for the backend.

After four complete rewrites and a heavy focus on documentation, I teased the launch on X with a simple date: "September 28." Despite having a small following, the post gained traction.

// The philosophy of Better Auth: Simple core, powerful plugins
const auth = betterAuth({
    database: myDb,
    plugins: [
        organizationPlugin(), // Add multi-tenancy easily
        twoFactorPlugin(),    // Add security layers
    ],
});

The Road to 1.0

The launch on September 28 sparked immediate interest. Users flooded the Discord and GitHub. The project remained in beta for three months, during which I followed a strict shipping schedule:

Weekly Friday releases (Sept - Nov 2024)
Continuous bug fixes and feature additions
Community-driven API refinements
Hit Version 1.0

The developer community took over, creating tutorials and blog posts, which accelerated our growth.

🏢 Transitioning to a Company

I had always studied Y Combinator (YC) videos, but I waited to apply until I had a proven product. I wasn't entirely sure how to monetize it yet, but I believed YC was the best place to figure that out.

I interviewed with Pete, and he took a chance on me. For someone born and raised in Ethiopia, getting into YC was a life-changing moment. Moving to San Francisco was a surreal experience:

"SF is basically X, but in real life."

Following YC, we secured funding led by Peak XV (formerly Sequoia India Southeast Asia), with contributions from over 50 other angel investors and funds. This allowed me to think about the bigger picture: hiring, company culture, and the long-term evolution of the software.

🔄 The Full Circle: Acquiring Auth.js

Later, in conversations with Balázs regarding the future of Auth.js, it became clear that our visions were aligned. We saw an opportunity for Better Auth to take over the project.

The Strategy:

Maintain Auth.js and NextAuth.js to fix security bugs.
Provide a long-term migration path for users to move to Better Auth.
Consolidate the ecosystem.

This was a poetic moment for me, as NextAuth was the very tool that inspired the creation of Better Auth. As a bonus, we now own the auth npm package! :)

🗺️ The Evolution Path

Better Auth

What's Next?

As we integrate with Vercel, we will continue to push the boundaries of what open-source authentication can be, ensuring that whether you are building a simple blog or a complex AI-driven enterprise platform, your auth is seamless and secure.