Designing Machine Learning Systems (Chip Huyen) — The most practical book on ML system design. Covers data engineering, feature stores, model deployment, monitoring, and infrastructure.
Building Machine Learning Pipelines (Hannelore Hämäläinen & others) — End-to-end guide to building ML pipelines with TFX, covering data validation, transformation, and orchestration.
Machine Learning Engineering (Andriy Burkov) — Production ML from a practical engineering perspective. Covers serving, monitoring, and team processes.
Designing Data-Intensive Applications (Martin Kleppmann) — Foundational reading for understanding the distributed systems concepts underlying AI infrastructure.
AI Engineering (Chip Huyen, 2025) — The 2025 successor covering LLMOps, agentic AI, MCP, and modern AI system patterns.
Online Courses
ML System Design (Stanford CS 329S) — Chip Huyen's Stanford course. Slides and resources freely available.
Full Stack Deep Learning — Comprehensive course on production ML: infrastructure, deployment, monitoring, and project management.
Made With ML — Practical MLOps tutorials, design patterns, and production-ready code.
DeepLearning.AI: ML Production Engineering — Coursera specialization on production ML systems.
Google Cloud MLOps — Hands-on labs for building ML pipelines on GCP.
Certifications
Certification
Provider
Focus Area
ML Engineer — Google Cloud
Google
ML pipelines, feature engineering, model serving on GCP
Netflix TechBlog — Recommendation systems, content understanding, ML infrastructure
Uber Engineering — ML at scale, Michelangelo platform, real-time predictions
DoorDash Engineering — ML for delivery logistics, ranking, ETA prediction
Spotify Engineering — Music recommendation, audio features, personalization at scale
Pinterest Engineering — Visual search, content discovery, ML pipelines
Meta Engineering (AI) — Large-scale ML, recommenders, NLP, computer vision
Exercise: Create your personalized learning roadmap. Pick one book, one course, and one certification from the lists above. For each, write down: a) your expected completion date, b) the key skill you expect to gain, c) how you will apply it (build a project, contribute to open-source, write a blog post).