Finally got Rednote scraping to work reliably after my friend dragged me into it
A friend of mine needed Rednote data for competitor monitoring and research, knew I build scrapers on the side, and basically guilt-tripped me into taking a look. Most of the existing scrapers were flaky and kept breaking his workflow so he figured I could do better.
Took a crack at it and yeah, it is not easy. Xiaohongshu blocks datacenter proxies almost immediately and the anti-bot system is genuinely one of the more aggressive ones I have dealt with. Did some things differently that I will keep to myself, but eventually got it to a point where it pulls accurate, consistent data every single run.
It scrapes profiles, bio, followers, likes, RedID, location, tags, and also posts with full engagement data. The one downside is cost, residential proxies are not cheap and that is baked into the price. But a scraper that actually works beats a cheap one that fails half the time.
If you want to check it out, links are in the comment.