QuestionQ68

VerifiedServing and scaling models

You work for a gaming company that creates massively multiplayer online (MMO) games. You built a TensorFlow model that predicts whether players will make in-app purchases of more than $10 during the next two weeks. The model’s predictions will be used to tailor each user’s game experience. User data is stored in BigQuery. How should you serve the model while optimizing cost, user experience, and ease of management?

Explanation

A streaming Dataflow pipeline can consume each in-app purchase event from Pub/Sub, run TensorFlow inference within the pipeline, and persist the updated prediction to Cloud SQL for the game application. This provides timely event-driven personalization while keeping inference centrally managed and avoiding per-device model distribution. Dataflow ML supports TensorFlow inference in streaming pipelines through Apache Beam’s RunInference transform.

Learn more

Community Discussion

No comments yet. Be the first to start the discussion!