A team is designing an AI assistant that helps users with travel planning.

The NVIDIA implementation angle is not cosmetic here: long-running agents should retrieve compact relevant context instead of replaying the entire conversation history into every call. Travel personalization depends on persistent preferences and multi-step plan updates. A single-turn answerer cannot adapt itineraries as constraints change. From an NVIDIA systems-engineering lens, Option C aligns with the way agentic services should be decomposed and measured. The selected option specifically C states “Engineering multi-step reasoning frameworks with persistent memory systems to store and utilize user preferences.”, which matches the operational requirement rather than a superficial wording match. The correct implementation surface is checkpointed state keyed by session or user, with schemas that preserve only the fields the workflow needs later. The losing choices mostly optimize for short-term convenience; unbounded memory creates privacy, relevance, and performance problems unless persistence is deliberate. This choice gives engineering teams the knobs they need for continuous tuning after deployment. The memory policy should define what is persisted, what is summarized, and what is discarded to avoid both context loss and prompt bloat.