Amazon Personalize
Personalized recommendations
Build recommendation systems using the same technology as Amazon.com: you provide user–item interaction data, it trains models and returns recommendations through an API.
- Cost$$ Medium
- Difficulty●●●●●
- Costs money when idle?Yes
- Managed by AWS
- In examsCLF · SAA
Think of it asLike a salesperson who remembers each customer's tastes: "you usually buy coffee, why not try this new bean?"
When to use it
✓ Use it when
You need per-user product or content recommendations without your own ML team.
✗ Avoid it when
Simple rule-based recommendations (best sellers) or you need full control over the algorithm → build your own / SageMaker AI.
Common uses
- Product recommendations on e-commerce sites
- Re-rank video and article content
- Personalized email marketing
What does it cost?
- Pricing model: Charged by data ingested, training hours, and number of recommendations (real-time by throughput, or batch).
⚠️ Cost traps
A created campaign is still charged for minimum capacity even if nobody calls it.
