I built YTDb, an “IMDb for YouTube” to help you discover channels by vibe instead of whatever the feed decides. It’s in open beta:
https://ytdb.me
Why:
Search and recommendations often feel like the same loop. I wanted a manual, people-first way to find creators that match the style you like (e.g., chill vs. high-energy, casual vs. scholarly) and to surface hidden gems outside the algorithm.
How it works:
Viewers can leave an overall rating (1–10) and adjust style sliders (pace/energy, polish, humor, depth, etc.).
Optional written reviews (75+ words) encourage thoughtful, non-spammy input.
Search by topic + vibe and optionally weight/ignore sliders; add filters like min subs, average video length, upload cadence.
Results return a match % to your configuration and “similar creators” based on style.
What I’m looking for from HN:
Does the onboarding make sense?
Are the slider dimensions useful/complete? What would you add/remove?
How do the matches feel vs. your expectations? Any obvious ranking misses?
What would make you willing to contribute reviews so the index improves?
Known limitations (beta):
Early dataset → some niches are thin; quality improves as real people add reviews.
Google sign-in required today (to reduce spam); exploring guest/alt logins.
Desktop-first; mobile polish in progress.
Near-term roadmap:
Guest read-only mode and non-Google login.
Creator pages (claim/verify), public lists (“If you like X, try Y/Z”), and “similar-to” pages.
Better niche taxonomies, smarter weighting, and exportable watchlists.
Community hub (Discord) for weekly “Channel of the Week.”
If this scratches an itch, I’d love your feedback (and a couple of reviews to seed the DB). Thanks for reading!
Hi HN —
I built YTDb, an “IMDb for YouTube” to help you discover channels by vibe instead of whatever the feed decides. It’s in open beta: https://ytdb.me
Why:
Search and recommendations often feel like the same loop. I wanted a manual, people-first way to find creators that match the style you like (e.g., chill vs. high-energy, casual vs. scholarly) and to surface hidden gems outside the algorithm.
How it works:
Viewers can leave an overall rating (1–10) and adjust style sliders (pace/energy, polish, humor, depth, etc.).
Optional written reviews (75+ words) encourage thoughtful, non-spammy input.
Search by topic + vibe and optionally weight/ignore sliders; add filters like min subs, average video length, upload cadence.
Results return a match % to your configuration and “similar creators” based on style.
What I’m looking for from HN:
Does the onboarding make sense?
Are the slider dimensions useful/complete? What would you add/remove?
How do the matches feel vs. your expectations? Any obvious ranking misses?
What would make you willing to contribute reviews so the index improves?
Known limitations (beta):
Early dataset → some niches are thin; quality improves as real people add reviews.
Google sign-in required today (to reduce spam); exploring guest/alt logins.
Desktop-first; mobile polish in progress.
Near-term roadmap:
Guest read-only mode and non-Google login.
Creator pages (claim/verify), public lists (“If you like X, try Y/Z”), and “similar-to” pages.
Better niche taxonomies, smarter weighting, and exportable watchlists.
Community hub (Discord) for weekly “Channel of the Week.”
If this scratches an itch, I’d love your feedback (and a couple of reviews to seed the DB). Thanks for reading!
Site: https://ytdb.me
Discord: https://discord.gg/86bVx8s2
Email: ytdbcontact@gmail.com