A reading list for the ML engineer who actually cares about systems

What it does

This repository is a curated index of research and resources at the intersection of machine learning and systems engineering. It catalogs papers, books, video lectures, and courses across categories like data processing, training systems, inference serving, and domain-specific stacks for video, edge, and federated learning. A dedicated section tracks LLM training and serving infrastructure separately from general ML/DL concerns.

The interesting bit

The maintainers treat this as a living syllabus rather than a static bookmark dump. They include links to code implementations where available, flag must-read whitepapers, and even host video tutorials on YouTube and Bilibili. A team rather than a single maintainer now curates submissions via pull request templates.

Key highlights

• Coverage spans top-tier venues: OSDI, SOSP, NSDI, MLSys, SIGCOMM, SoCC, EuroSys
• Separate tracks for ML/DL infra and LLM-specific infrastructure (training and serving)
• Domain-specific sections for AutoML, GNNs, edge/mobile, video, and federated learning
• Curated course list includes Berkeley’s CS294, Stanford’s CS329S, and Tianqi Chen’s CSE 599W
• Companion website “Lets Go AI” in development

Caveats

• The README is essentially a table of contents; actual paper summaries and code links live in separate markdown files not shown in the provided source
• Some sections (blogs, books) appear lightly curated compared to the paper indexes
• No explicit criteria for inclusion or ranking of resources is stated

Verdict

Worth bookmarking if you’re building ML infrastructure and want a structured path through the systems literature. Skip it if you need executable code or hands-on tutorials; this is a map, not a toolbox.