THE STATE OF MLOPS - MACHINE LEARNING IN PRODUCTION AT ENTERPRISE SCALE
summary
Argues MLOps is DevOps applied to ML, with concrete architecture: Airflow training jobs produce serialized models stored in a model repository + Kubernetes serving flow used at Wayfair; alternatives are embedding models, a shared model service, per-model microservices, or in-memory streaming for low latency. Recommends monitoring both technical and business metrics, using feature stores for precomputed features, and choosing tools to fit the organization — examples named: SageMaker, Vertex AI, Azure ML, Databricks.
AI-written summary · about
chapters (11)
- 0:00Introduction
- 2:07What is ML
- 14:48Technology
- 18:08Architecture
- 21:55Example
- 23:51Tool Selection
- 26:24Running Models in Production
- 32:51Feature Store
- 35:13Recap
- 36:02Resources
- 36:37Questions
related talks
47mMachine Learning in Production: An Intro to MLOps.
31mMLOps In Practice – How To Run Your Machine Learning Models In Production At Enterprise Scale
45mThe Culture, Architecture and Technology of MLOps
13mMLOps at Scale
01mMLOps: Applying DevOps to ML
53mExperience with an end-to-end MLOps pipeline: Practical insights and 3D architectural perspectives
13mMLOps: From AI Research to Production-Ready Systems
20mBringing CI/CD Practices to Machine Learning with MLOps