Dev builds Chinese learning platform with 100k+ TV clips using Rust and Next.js

A developer created MandarinClips, a language learning platform indexing over 100,000 sentence-length clips from Chinese TV shows. The backend uses Rust with Axum for low-latency search, and PostgreSQL trigram indexes enable sub-millisecond fuzzy matching. Video hosting on Backblaze B2 via Cloudflare eliminates bandwidth costs.