The most intuitive explanation of GPs came to me from McElreath's Statistical rethinking. What was most sticky was that was explained in terms of a thing needed

The most intuitive explanation of GPs came to me from McElreath's Statistical rethinking. What was most sticky was that was explained in terms of a thing needed for a simple regression problem that becomes hairy with continuous variable. It was an example of modeling a caffe's wait time which varies more continuously with different covariates like time of the day and IIRC size of the cafe.