A media company is building a content moderation pipeline using the Anthropic Batch API. Each batch contains 10,000 image captions to classify. The pipeline currently polls the batch status endpoint every 5 seconds. During peak hours the team observes 429 errors on the polling endpoint. What is the correct architectural response?

B is correct because polling at any fixed interval creates unnecessary load on the status endpoint and is the root cause of 429 errors during peak hours. The Batch API supports webhook callbacks that notify the caller when processing is complete, eliminating polling entirely. The webhook pattern is event-driven and zero-load between batch submission and completion. C is wrong because reducing the polling interval to 1 second dramatically increases the number of status requests, making 429 errors worse rather than better. A is wrong because 429 errors on the polling endpoint are caused by request frequency, not by batch size or max_tokens; increasing per-request token counts does not affect polling rate limits. B is wrong because canceling and resubmitting to a secondary key does not solve the architectural issue of polling-induced rate pressure and risks losing work; it is a workaround, not a solution.