A syllabus for the self-taught ML engineer

What it does This repo is a curated index of publicly available machine learning courses from top universities—Berkeley, MIT, Stanford, Columbia, Caltech—plus a few industry case studies. It maps out a full-stack path: computer science fundamentals, math, core ML theory, deep learning, and specializations from NLP to robotics. The maintainer marks foundational courses with :star: and links to video lectures where available.

The interesting bit Instead of dumping links, the README structures learning as decision trees. There’s a “Shortest Path to LLM / Agents” for the impatient, a “TL;DR” for basic ML engineering literacy, and then deep dives by topic. The maintainer also slips in their own reference solutions for some courses (CS182, CS224n), suggesting this is a living study guide, not just a bookmark list.

Key highlights

• Covers the full stack: CS fundamentals → math → AI/ML theory → engineering → specializations (vision, NLP, RL, robotics, foundation models)
• Includes practical infrastructure courses like Berkeley’s “Full Stack Deep Learning” and MIT’s “Missing Semester”
• Curated textbooks alongside courses (ESL, PRML, Bishop, etc.)
• Industry case studies section for applied context
• CC0 license, actively maintained (last-commit badge visible)

Caveats

• Some sections in the README appear truncated (the “Machine Learning” courses list cuts off mid-entry)
• No clear indication of how often course links are validated—university course pages rot
• The “awesome list” format means minimal original content; value is entirely in curation

Verdict Worth bookmarking if you’re self-studying ML and tired of forum threads arguing about which Stanford course to take first. Skip it if you already have a curriculum or want depth beyond syllabi and video links.