The model already found titles that score higher. Try one.
This is a small HN-like sandbox for testing a submission before posting it for real. I trained a model on ~150k HN submissions to estimate score, front-page odds, and likely range. The useful part for me is not the exact number, but seeing similar past posts and why the model thinks a title is weak or strong.
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.