About MoodRoulette

The problem: streaming choice paralysis

Most evenings don't fail because there's nothing to watch — they fail because there's far too much. Four subscriptions, thousands of titles, a group of people with different tastes, and forty minutes of scrolling later everyone gives up and puts on something they've already seen. MoodRoulette exists to end that negotiation in about thirty seconds.

Who builds it

MoodRoulette is designed and maintained by a working software engineer who kept watching the same argument play out on the sofa and decided to treat it as an engineering problem rather than a taste problem. The tool is built with the same practices used in production software: a pure, testable filtering engine separated from the interface, a real-time multiplayer layer backed by Postgres, server-side caching of catalogue data, and region-aware availability so a recommendation is only shown if you can actually press play on it.

Every product decision comes from using the thing in real life — with real subscriptions, real friends and real Friday nights — and the site is iterated on continuously based on that experience and on feedback sent in by users.

How it works

• Mood Roulette (solo). Pick your country and streaming services, choose a vibe, set runtime, year and rating limits — and get one confident pick rather than an endless grid.
• Swipe Night (group). Create a six-character room code, everyone swipes on the same deck from their own phone, and the moment every person says yes to the same film you all get an instant match.
• Only what you can stream. Results are filtered against the services you actually pay for in your country, using live catalogue and availability data.

How we stay honest

MoodRoulette is free to use. It's funded by advertising and by affiliate commissions on outbound "Watch Now" links, which never change which film is recommended to you — the ranking is driven only by your filters. We're not paid by any streaming service to favour its catalogue, and we say so plainly in our Privacy Policy. If a recommendation or availability listing is wrong, please tell us — corrections get shipped quickly.