The correct answer is B because Intellisense in Azure Machine Learning Studio notebooks requires a running compute instance to provide code completion and inline error highlighting. The community discussion shows strong consensus for option B with multiple upvoted comments explaining that Intellisense depends on the compute instance being active. Options C and D (%pip and !pip magic functions) are incorrect because they are used for package installation, not for enabling Intellisense functionality. Option A (stopping the compute instance) would actually disable Intellisense rather than enable it. Microsoft documentation confirms that when a compute instance is running, you can use code completion powered by Intellisense in Python notebooks.