I don't think the hypothetical people you refer to would be against flagging people for bigoted behavior/speech. In fact, I personally don't see an issue with m

I don't think the hypothetical people you refer to would be against flagging people for bigoted behavior/speech. In fact, I personally don't see an issue with manually doing a Twitter advanced search to see if someone – especially for a public-facing role – has tweeted outright slurs; I specify "manually" to insinuate that context is important when assessing this. I think people's main issues with this specific software/practice are: 1. It evaluates Liked tweets – if you believe tweets to have too little context to be accurate signals of how a potential employee would behave, liking a tweet contains even less context. 2. Any service that provides data via widescale scraping and auto classifying can and will eventually be reduced to an "authoritative" score.