Personalisation at scale with AI/ML and Amazon Personalize
summary
Pipeline ingests user interactions, item metadata and user metadata into Snowflake, streams events via Kinesis/API Gateway to call Personalize PutEvent for online updates, and serves recommendations through a Personalize campaign into Braze for newsletter placement. Implementation uses serverless Lambdas, AWS Data Lab engagement and built‑in NLP and exploration controls in Amazon Personalize. Results show 250% uplift in click-through conversion, 49% more tickets sold per email opened, and ~5x greater catalog diversity, proving the end-to-end system works at scale.
AI-written summary · about
chapters (9)
- 0:00Introduction
- 0:56Value of personalization
- 5:05Recommender system approaches
- 10:05Amazon personalize overview
- 14:04Technical architecture
- 15:43Ticketek case study
- 21:30Implementation best practices
- 23:41Results and future outlook
- 25:05Getting started with AWS
related talks
29mPioneering user experiences at StockX with Amazon Personalize
25mDeliver viewing experiences for super fans with Amazon Personalize
27mTransform online shopping experiences with personalized recommendations
27mHow Warner Bros. Discovery uses Amazon Personalize
11mMaximizing personalization and AI models with quality customer data
24mAI-Powered Transformation in Fashion E-Commerce
28mThe Power of Personalized Customer Experiences Through AI Innovations
18mBehavioral Data Creation for AI - Delivering Value and Insights