Browser-Based Video Editing Meets Vercel
Ozone, a collaborative video editing platform that runs entirely in the browser, made a strategic infrastructure change that redefined its development workflow. The platform—which leverages React and WebAssembly to handle heavy rendering tasks and large datasets—moved from Heroku to Vercel to address chronic slow deployments and service interruptions.
The effect was immediate. Deployment tasks that previously consumed 15-20 minutes were reduced to seconds, accelerating the overall development process by fivefold. Build times dropped by 85%, a meaningful improvement for a product with many custom submodules and demanding client-side rendering requirements.
Independent Feature Delivery
The switch unlocked a more collaborative engineering culture for the small Ozone team, which juggles hundreds of features across areas like AI, timeline editing, and export functionality. With Vercel's deployment model, each feature can be pushed to its own unique branch with an accompanying preview URL, and teammates can leave comments directly on those branches.
This removes cross-team bottlenecks. One engineer can be testing an AI integration while another is iterating on the timeline, and neither has to wait for the other to finish before validating their work in a production-like environment.
"Knowing that an engineer can finish working on something that I can test in staging or production within a matter of what seconds is incredible. In the past, I was waiting 15 to 20 minutes to be able to get my hands on anything."
![]()
Max von Wolff, Founder & CEO
AI Built Into the Editing Flow
Ozone is now integrating generative AI into its platform, letting users create video, images, music, and speech on demand by connecting to models from providers such as Stability AI, Meta, and ElevenLabs. The move aims to streamline creative workflows so users can generate assets without leaving the editing environment.
The Vercel ecosystem eased this adoption curve. The team used the AI SDK to build streaming conversational interfaces and leveraged AI Integrations from the Vercel dashboard to wire external model APIs directly into the codebase, test them, and push to production.
"It would be foolish not to embrace this incredible AI trend that is happening. And with Vercel, its easy to allow creators to really let their creativity run wild and create it all in one platform by incorporating AI easily into our platform."
![]()
Alexander Kollmann, Product Manager
Iteration as a Competitive Edge
For Ozone, the ability to quickly turn user feedback into shipped features is as much a business advantage as a technical one. The faster feedback loop made possible by Vercel's infrastructure lets the small team tackle a large roadmap while keeping up with evolving user expectations.
With development cycles shortened and the barrier to AI integration lowered, Ozone sees a continued path toward expanding what creators can accomplish inside a browser tab.



