This is a dedicated watch page for a single video.
You are building a recommendation engine on Google Cloud that suggests new videos to users based on their past viewing behavior. Your solution must generate labels for the content in videos the user has already watched and use those labels to make fast, personalized suggestions. The system must handle several terabytes of data and quickly filter results based on patterns observed in other users’ preferences. Which approach should you take to meet these performance and scalability goals?