Todd Underwood

Books in this wiki: Reliable Machine Learning

Background

Site Reliability Engineer and ML systems practitioner. Co-author of Reliable Machine Learning (O'Reilly, 2022). Brings deep SRE expertise to ML production operations.

Core Positions

The SRE discipline — SLOs, error budgets, monitoring taxonomies, progressive rollouts — applies directly to ML systems with some ML-specific additions around model quality signals and training pipeline reliability.