HN Showcase
Back to browse
Tastefinder – swipe-based movie and TV recommendations

Tastefinder – swipe-based movie and TV recommendations

by tastefinder_io·Feb 21, 2026·2 points·0 comments
Visit ProjectView on HN

AI Analysis

●MidEye CandyCozy

Tinder for movies, but Letterboxd, Netflix, and JustWatch already own this problem better.

Strengths
  • •Swipe UX is genuinely delightful and faster than filter menus; mobile-native feel translates well to web.
  • •Multi-strategy recommendation engine (user-based, content-based, trend-based) shows real thought beyond simple filtering.
  • •Guest mode cold-start is thoughtfully designed to avoid login friction on first discovery.
Weaknesses
  • •Recommendation quality is unproven; no public benchmarks against Netflix/JustWatch baseline accuracy.
  • •Fundamental problem ("what should I watch?") is already solved by Netflix Browse, Letterboxd, and JustWatch with richer metadata.
Category
Productivity
Target Audience

Casual streamers seeking effortless content discovery; users who prefer mobile-like interaction over scrolling lists.

Similar To

Letterboxd · Netflix Browse · JustWatch

Post Description

I built Tastefinder to make “what should I watch?” less painful.

It’s a card UI with 4 reactions: * right = like * left = dislike * up = super like * down = skip

Signed-in users get recommendations from a full multi-strategy engine Guests use a lighter recommendation path based on their current reactions.

You can also filter by type, genre, country, year, IMDb, and Rotten Tomatoes.

I’d love feedback on recommendation quality, cold start behavior, and swipe UX.

https://tastefinder.io

Similar Projects