Collaborative filtering makes recommendations from interaction patterns alone. It never looks at what an item is — only at who engaged with it. That is simultaneously its great strength and the source of its cold-start failure.
Collaborative filtering cannot recommend an item nobody has touched. Content-based filtering can, because it looks at what the item is rather than who engaged with it — and every production system ends up combining the two.
Recommender evaluation is unusually treacherous: the offline metric is computed on data generated by a previous model, the ranking metrics disagree with each other, and the correlation between offline improvement and online business impact is famously weak.
Recommendation looks like supervised learning and behaves nothing like it. The data is almost entirely missing, the missingness is not random, the output is a ranked list rather than a prediction, and the model's own outputs determine what data you collect next.