Applied AI·Production patterns and cost
after enough consecutive failures you stop calling the provider at all for a while, instead of queueing thousands of doomed requests behind a dead dependency.
Circuit breaker
Draft summary, pending review
After sustained failures or latency spikes from a provider, stop calling it for a cooling period and serve fallbacks, rather than queueing doomed requests. Your existing resilience pattern, unchanged.