Hi HN,
Podcast discovery feels like roulette — some episodes fire me up, others drain me.
I built Qling to rank shows by inspiration and merit, with previews to give bite sized highlight snippets.
How it works:
– The ingestion engine ingests ~20k new episodes/week
– Transcripts: Whisper-large-v3-turbo and Meta-Llama-3-8B-Instruct on an A6000 GPU farm
– Scoring: episode embeddings + speaker credibility + insight + inspiration
– Stack: Django + Postgres + Redis on Heroku; SwiftUI client
Try it:
- iOS app: https://apps.apple.com/app/qling-podcast-player/id6740029729
- Simple landing page: https://www.qling.app/
Looking for feedback on:
1. Do highlight snippets help you decide faster?
2. Are the “inspiring” picks accurate or cheesy?
3. Any UX issues?
I’ll be in the thread all day — ask me anything and I’ll ship fixes asap.
Thanks for checking it out!
Comments URL: https://news.ycombinator.com/item?id=44404792
Points: 1
# Comments: 0
.png)
