Personalization in 10 Milliseconds
summary
Tinkoff.ru describes moving from classical A/B testing to per-user page personalization using machine learning to improve online customer acquisition; the talk covers using ML algorithms to customize web pages for individual users and the engineering needed to deliver personalized pages with low latency (10 milliseconds).
AI-written summary · about
chapters (12)
- 0:00Introduction
- 3:13A/B testing issues
- 7:25System requirements
- 11:36First results and problems
- 16:33MVP architecture problems
- 19:18Pipeline architecture
- 22:48Pipeline implementation
- 27:01Machine learning integration
- 29:10Results and performance
- 32:42Business cases and benefits
- 35:10Conclusion and recommendations
- 35:55Q&A
related talks
19mMachine Learning is a new age of Personalized Game Monetization
03mMachine Learning for Personalization
31mA Feasibility of On-Device ML for Adaptive User Interfaces in Frontend Development
15mPersonalization, Performance, and Probably Dynamic Content
26mPersonalisation at scale with AI/ML and Amazon Personalize
27mTransform online shopping experiences with personalized recommendations
39mGrowing user engagement with RL-driven personalisation
22mAsynchronous Personalization