#ai-operations

The most expensive AI agent failures don't throw an error, they hide. One ran at a six-figure-a-year rate for days while every dashboard stayed green, because the signals that catch it, per-session cost and anomalies, are the ones nobody watches. Why agent loops run away, and the two cost instruments your monitoring is missing.

Strands' BedrockModel ships with prompt caching off. Two kwargs turn it on, one per-model gotcha catches you, and a 10-turn driver measures 99.9% / 99.8% hit ratios on Nova Pro and Sonnet 4.6 against an 8,156-token production system prefix. The per-call usage block proves it in seconds, not waiting on CloudWatch.

Three findings from a real customer engagement on AWS Bedrock: what a load test was actually doing, why p95 latency was 45 seconds, and the prompt-caching default that costs every team money. Plus the three CloudWatch metrics that catch all three.