Real-time Machine Learning: Architecture and Challenges
- Use cases for real-time ML.
- Architectures well suited for online predictions taking feature computation, prediction, and request response times into consideration.
- How to overcome key challenges with online prediction systems such as latency vs. feature freshness, accuracy, and streaming infrastructure management.
Co-founder of Claypot AI
About the presenter
Chip Huyen is a co-founder of Claypot AI, a platform for real-time machine learning. Previously, she was with Snorkel AI and NVIDIA. She teaches CS 329S: Machine Learning Systems Design at Stanford. She’s the author of the book Designing Machine Learning Systems (O’Reilly, 2022).