A data science team needs to enable real-time machine learning inference for a dynamic pricing model in a retail environment. The solution requires minimal latency and the ability to scale automatically based on traffic. Which AWS service should they use?