Q: 14
Scenario: An administrator must ensure that each AI developer can access only their assigned SageMaker notebook instance while maintaining shared access to Amazon Rekognition APIs and training data stored in Amazon S3. Question- Which solution will meet this requirement?. Options:
Options
Discussion
Option A
A. since IAM policies can lock down SageMaker permissions by notebook ARN for each user. B is a trap here, because JupyterLab roles don't cover AWS-level access controls. Pretty sure A is best, but open to counterpoints.
I thought D might work since lifecycle configs can block some access at notebook start. Maybe not as tight as IAM policies, but it can help restrict what happens on unassigned notebooks. Anyone else choose D for scenarios like this?
Saw something like this in a practice test, it's D. When Clarify says there's significant skew, it's almost always class imbalance causing poor minority class detection.
AWS always asks this kind of class imbalance! D
A is the way to go here since IAM lets you lock down SageMaker by notebook ARN per user, which covers the isolation part. Pretty sure that's what AWS expects because it follows least privilege. Could see confusion with D but think A matches exam intent, agree?
C blocks network but not actual notebook instance access, so that's a common trap. I don’t think B solves this because JupyterLab roles don’t enforce AWS resource limits. A fits best since IAM restricts by notebook ARN for each dev. Open if someone thinks D is better, but pretty sure A is what AWS exams want.
A , matches what the official AWS guide and practice questions focus on for least privilege. If you want to double check, try the official documentation or exam labs for IAM + SageMaker restrictions.
A that's the IAM way. Least privilege per user is what AWS expects for this kind of control.
Had something like this in a mock, it's A. Specific IAM policies per developer let you lock access to only their notebook instance and still give shared Rekognition/S3 rights. Anyone else pick A here?
Be respectful. No spam.
