Manish Gupta: Director, American Express at Cypher 2017 - Part 1/3
summary
Built end-to-end personalization for digital commerce that serves real-time personalization in milliseconds over hundreds of millions of records, using NLP/LSI, collaborative filtering and semantic mappings to handle 1.5M daily applications and 4–5M daily searches across ~300k live jobs. Deployed on a 24-core / 128GB production node and scaled pipelines for 270M rows and millions of CVs. A contextual course-recommendation experiment increases revenue fivefold (from $4M to $20M) and demonstrates that personalized recommendations plus semantic search drive conversion.
AI-written summary · about
chapters (5)
- 0:00Start
- 2:06Why Do We Need Data Signs for Digital Commerce
- 14:14Personalized Recommendation Systems
- 17:00Semantic Search
- 22:50Contextual Advertising
related talks
27mTransform online shopping experiences with personalized recommendations
28mCrafting Hyper-Relevant User Experiences
26mPersonalisation at scale with AI/ML and Amazon Personalize
29mPioneering user experiences at StockX with Amazon Personalize
22mAI-Powered Personalized Discovery with Graph Search and VLM-Enhanced Embeddings
28mThe Power of Personalized Customer Experiences Through AI Innovations
31mAdvanced search for your e-commerce website
20mDataEngBytes 2023 - PER-T1-06 - Kamal Hossain