The model already found titles that score higher. Try one.
I analyzed around 150k Hacker News submissions and built a small tool that lets you test a title, URL, and text before posting. It predicts likely score, front-page probability, and shows similar historical posts so you can understand what the model is comparing against. I built it because Show HN launches often die without any feedback, and I wanted a way to debug the title before hitting submit.
ForesynWanna keep in touch?
Built this solo over a weekend. Soft-launching before the HN post on Monday. If you scored a draft and the prediction either nailed it or whiffed, I want to know.