The End of Glitch (Even Though They Say It Isn't)

May 23, 2025 blog glitch feature

The news is out! Glitch is shutting down project hosting and user profiles on July 8th, 2025. Like Jenn, I knew this was coming because working on a team for five years gives you the inside scoop, even after you’ve been gone for a year.

I don’t envy Anil’s job in writing that blog post, or any of my incredible former teammates who are still at Fastly working to make this transition as painless as possible. But let’s be real about what this is.

The post carefully avoids calling this an “Our Incredible Journey” moment, but removing project hosting and user profiles is the end of Glitch as a platform. What’s left is essentially a redirect service with some backlinks (hosted on the Fastly Edge Compute Platform™️, naturally). You can’t spin the removal of the core product as anything other than what it is: a shutdown.

What Made Glitch Special

What brought me to Glitch was a combination of factors that felt too good to pass up. The team was incredibly diverse and genuinely cool—the kind of people who cared deeply about making the web more accessible and creative. The opportunity was massive: building something that people used daily, helping deploy millions of apps, and getting to shape what the future of web development could look like. And Glitch had this unique position in the industry where we constantly punched above our weight—getting media attention, making statements that got picked up across tech, having industry eyes on everything we built.

But what kept me there for five years were the people and the projects our users built daily. Getting a tweet from the BTS Army—who built one of the most popular apps on the platform to track Spotify charts for every single BTS song—and watching thousands of fans jump on and start tweeting at us was a career highlight. It was nerve wracking every time we had a big release, but the excitement and joy from our users made it all worth it. The community was a huge part of what made Glitch special. We had a vibrant ecosystem of creators, artists, and developers who were all there to learn and share.

Glitch was magical because you could go from idea to deployed app in minutes. No build processes, no deployment pipelines, no credit card required. Just start typing and your code was live on the internet. For tons of people, it was their first taste of what it felt like to be a real developer.

The remix culture was unlike anything else. See a cool project? Hit remix and make it your own. The platform encouraged tinkering, learning, and sharing in ways that felt genuinely collaborative rather than competitive.

Glitch in Bio: Everyone Gets a Website

One of my favorite launches was Glitch in Bio — a dead-simple way for anyone to spin up a personal website. We watched people discover they could have their own corner of the internet, not just another social media profile. When Bluesky started gaining traction, we made it trivially easy to use your Glitch subdomain as your handle. Suddenly you had yourname.glitch.me as both your website AND your social identity.

This wasn’t just about making websites easier (though we did that). It was about democratizing the web. Every person deserves to own a piece of the internet, not just rent space on someone else’s platform, or paying $5 a month to post links to your other homes on the web.

What Went Wrong (Before Fastly)

Here’s the thing most people don’t know: Glitch started as a hack project by other employees at Fog Creek Software (which eventually became Glitch the company). It was brilliant in its scrappiness—a working prototype that captured the magic of instant deployment and collaborative coding. But it launched in that same prototype state.

By the time I took over engineering, we were constantly at war with our own technology. Every decision made sense when it was a quick hack for a few dozen users, but when you’re serving millions of apps? Those early architectural choices became technical debt that compounded daily. We’d fix one scaling issue only to discover three more hiding underneath.

The containerization system was held together with what felt like digital duct tape. The editor had performance issues that got worse as projects grew. The database groaned under the weight of all those projects and remixes. We kept trying to rewrite core systems while keeping the platform running—like changing the engine of a car while driving down the highway.

We didn’t move fast enough to address the fundamental infrastructure problems. Every month we delayed the big rewrites, the harder they became. The platform that made it effortless for users to create was becoming increasingly difficult for us to maintain.

Enter Fastly: The Cavalry Arrives

When Fastly acquired Glitch in 2022, it felt like salvation. Here were people who actually knew how to build massive-scale internet infrastructure! They’d solved the exact problems we were drowning in—running millions of compute instances, handling global traffic, dealing with abuse and security at scale.

I was genuinely excited. Fastly’s edge computing platform seemed like the perfect technical foundation for Glitch’s next chapter. Finally, we could focus on the creative tools and community features instead of constantly firefighting infrastructure issues.

That… didn’t work out as planned.

Without getting into the weeds (bridges to maintain and all), I’ll just say that the challenges of integrating a creative, community-focused platform into a more traditional enterprise infrastructure company proved harder than anyone anticipated. Different cultures, different priorities, different definitions of success.

The official post mentions that “it takes a lot of time and money to run millions of apps” and points to abuse issues. While both are true, that’s not the whole story. Glitch was never going to be a cash cow, but it didn’t need to be. It needed to be valued for what it was: a place where the next generation of developers got their start.

The First Tech Union

One of the things I’m most proud of from my time at Glitch was watching my coworkers form the first recognized union in tech. As a manager when the organizing started, I got to witness the incredible work, sweat, and tears that went into building something historic. The Glitch Workers Union was groundbreaking—showing that tech workers could organize collectively and that companies could work constructively with unions.

When we moved to Fastly, we made the difficult decision to dissolve the union, but not before negotiating solid deals for everyone transitioning over. That movement didn’t die with our union though—it’s still growing bigger every day across the tech industry. Glitch workers proved it was possible, and others are following that path.

The platform is gone, but that legacy of worker organizing continues to ripple outward.

Where This Leaves Us

No place online lasts forever, and there isn’t a reason it should—a lesson I’m still learning. Glitch was a beautiful thing for a long time, and now we all need to find new homes for our projects and communities.

Here’s where I’d recommend taking your projects:

Val Town captures the closest thing to Glitch’s spirit. You build small, composable functions that can work together—an API here, a frontend there, maybe a cron job in between. Their AI tools are genuinely helpful because you’re working on focused, small-scope projects where AI can actually understand what you’re trying to do. An editor in the browser that builds and deploys your code instantly, who would have thought!

Netlify is the reliable choice for static sites. If you built an Eleventy or simple HTML site on Glitch, Netlify will feel familiar and probably perform better.

GitHub Pages for simple static hosting, especially if your project is already in a GitHub repo.

Fly.io if you want more control and don’t mind a steeper learning curve. It’s almost entirely CLI based, but it gives you a lot of flexibility and power. You can run anything from a simple web app to a full-blown microservices architecture.

What We Lost

The thing that breaks my heart isn’t just that another platform is shutting down—it’s that we’re losing one of the last places on the internet that prioritized joy and experimentation over engagement metrics and revenue optimization.

For a few years, Glitch was the closest thing we had to the early web’s spirit of “view source” and “anyone can build here.” But let’s be honest: we proved that you can’t build a sustainable platform around helping people create rather than consume—at least not in today’s internet economy. The math just doesn’t work when hosting costs real money and creative tools don’t generate the kind of engagement that sells ads or subscriptions.

If you’re migrating off Glitch and need help, the community forum is still active, and the team is committed to making the transition as smooth as possible. Take advantage of the code export tools—your work deserves to live on.