CT-AI Question 31

Community Discussion

Written by other learners. Comments can disagree with the marked answer — check the official documentation when in doubt.

• Chose C0

  - In a canary deployment, the new model serves a small slice of production traffic (e.g., 1-5%), with metrics monitored before scaling up. Shadow deployment runs the new model in parallel without serving its responses to users (useful for offline comparison). A/B testing splits traffic between two variants for statistical comparison. Blue-green keeps two full environments and switches traffic atomically.