The ML Lifecycle in Combat Conditions
summary
Covers practical lifecycle of machine learning in production, arguing that model training is only about a quarter of the work while most effort goes to data preparation, production infrastructure, deployment and monitoring. Discusses common gaps where ML communities focus on algorithms but neglect operational concerns, and outlines engineering practices to manage data pipelines, deployment workflows and monitoring for reliable ML systems.
AI-written summary · about
chapters (7)
- 0:00Introduction to ML lifecycle
- 1:08Model deployment challenges
- 8:57Solutions and approaches
- 19:41System architecture
- 34:27Implementation and tools
- 41:14Results and conclusions
- 50:15Q&A
related talks
43mML application lifecycle
30mChallenges of machine learning: The new ML lifecycle
30mThe State of Production Machine Learning in 2023
06mMachine learning is about your data and deployment
47mMachine Learning in Production: An Intro to MLOps.
29mDemystifying Machine Learning in Production: Reasoning about a Large-Scale ML Platform
17mContinuous deployment for ML: The new software development lifecycle
21mWhy is Production ML so Hard?