Launching today
ScreenCI
Product videos that automatically update on every release
36 followers
Product videos that automatically update on every release
36 followers
ScreenCI helps SaaS teams keep product videos in sync with their product. It turns E2E tests into videos. A coding agent can draft scripts from your codebase or live app, which can run automatically in CI. Anyone on the team can fine-tune videos in the browser, localize them into 70+ languages, and publish approved versions to static URLs that always serve the latest version.
Hey Product Hunt, we built ScreenCI because we kept running into the same problem: product videos go stale fast, especially for SaaS teams shipping constantly. We wanted a way to keep docs, demos, and walkthroughs accurate without turning every product change into another editing session.
ScreenCI treats product video like code:
- Stale videos fail the build: every video is a TypeScript file driven by Playwright, so CI catches outdated walkthroughs before users do.
- Docs that match every release: walkthroughs ship with the product, keeping customer-facing videos and screenshots aligned with the current UI.
- Localize with ease: turn one source video into 70+ languages with AI narration and subtitles for a more accessible experience.
- Let your agent write the first draft: point your coding agent at your codebase or live app, and ScreenCI drafts the walkthrough script for you.
- Edit via the browser editor or with your AI agent: anyone on the team can refine narration, swap voices, and update the video in whichever workflow fits best.
- Publish once: ship approved videos to embeddable static links that always serve the latest version.
If your team has ever been stuck re-recording outdated walkthroughs, we’d love to hear how you handle it today and what would make ScreenCI more useful for you.
Curious how this handles apps that rely on seeded test data or ephemeral environments, since flaky CI runs would make the resulting videos unpredictable. Does the agent regenerate the script after every deploy or wait for a trigger?
@zmra1131403 Great question!
The agent is the one you already have, such as Claude Code or Cursor. The generated video script itself is unaffected unless you change it via other dev tools. The video is Playwright E2E in your codebase, with some extra cababilities, which you can trigger 1. manually or 2. automatically in your continuous integration (CI) pipeline.
It is the same situation as with using Playwright E2E tests with ephemeral products, such as products which have AI responses etc. It is great if your product is abstracted in a way that you can create environment with the same seed every time, so for example the AI response would be the same every time. This helps with ScreenCI video automation, but also testability and product maintainability in general. In my experience dependency injection powerful tool for these kinds of products.
when a UI change only affects one screen in a longer walkthrough, does it re-render the whole video or just re-splice the affected segment? Curious because if it's full re-renders every time, the 70+ language narration could get expensive fast on a team that ships daily, versus something more incremental.
@galdayan Great question! ScreenCI today supports splicing per video and per language, so you can choose which videos and languages gets triggered (could be a rule in CI based on modified files etc).
But what we have been developing behind a feature flag is a way to split videos into "sub-videos", meaning you could have a long walkthrough recorded and rendered in parts, and the "super-video" on top would stitch the parts together. Works already in code, UI still needs to be figured out with these dependencies. This would mean performance improvements but also allow larger organisations to split ownership of parts to different teams (+ some group could maintain the overlay assets and potentially narration from web editor).
That is the vision, does it sound like a good solution in the situation you have?
There is a real need for this kind of service. At my previous e-commerce business, when we removed one product from the product packaging or added a new one, we would have needed this!
@henri_autio Exactly! That’s a great example of the problem we’re trying to solve. Even a small product or packaging change can make an otherwise good video outdated.
Our goal with ScreenCI is to make those updates much easier, so teams don’t have to re-record everything manually every time something changes.
Have you had a chance to try the anonymous free trial yet? :)
I can support if you have any questions or blockers with the product. Should be one AI prompt = localized product video script + recording out