AIF-C01 Fundamentals of AI and ML Practice Question
A company is deploying a machine learning model for real-time fraud detection. The model must have latency under 100ms. Which infrastructure choice is most appropriate?
⚠ Common exam trap
Test-takers frequently confuse batch transform with real-time inference, assuming that any SageMaker inference capability can serve low-latency requests, but batch transform is explicitly asynchronous and designed for high-throughput, not low-latency.
Answer choices
Why each option matters
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
Amazon SageMaker real-time endpoints
Amazon SageMaker real-time endpoints are designed for low-latency inference, typically in the tens of milliseconds, making them suitable for real-time fraud detection where latency must be under 100ms. They deploy a model behind a persistent HTTPS endpoint that auto-scales to handle incoming requests with minimal delay.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✓
Amazon SageMaker real-time endpoints
Why this is correct
SageMaker real-time endpoints keep the model loaded on persistent instances and return synchronous predictions with consistently low latency, satisfying the sub-100ms requirement. Batch transform and asynchronous inference introduce queuing or storage overheads unsuitable for immediate fraud decisions.
- ✗
Amazon EC2 with Deep Learning AMI
Why it's wrong here
A general-purpose EC2 instance with a Deep Learning AMI leaves you managing servers, scaling and endpoint serving yourself, so sub-100ms latency is not guaranteed. It is tempting because it offers full control over the environment, and it would be correct for custom training workloads or bespoke frameworks needing GPU instances.
- ✗
Amazon SageMaker batch transform
Why it's wrong here
Batch transform processes an entire dataset as an asynchronous job, so it cannot return per-request predictions within a 100ms budget. It is tempting because it is a managed SageMaker inference mode, and it would be correct for offline scoring of large stored datasets where throughput matters rather than latency.
- ✗
Amazon SageMaker notebook instance
Why it's wrong here
Notebook instances are interactive development environments for authoring and experimentation, not low-latency inference hosting; they cannot serve sub-100ms predictions. They would be the right choice for exploratory data analysis or prototyping before deployment. Real-time fraud detection needs a continuously available endpoint such as SageMaker hosting.
Go deeper
Related to this question
About these practice questions
One of 862 original AIF-C01 practice questions on Courseiva, each with a full explanation and wrong-answer analysis — not exam dumps or protected exam content. Learn why practice questions differ from exam dumps →
JA
Written by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
This AIF-C01 practice question is part of Courseiva's free Amazon Web Services certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the AIF-C01 exam.