The infrastructure choices behind getting predictions out of a trained model at production latency and scale.
Reference Q&A for this concept is coming soon.
