A curated roadmap of learning material for machine learning, artificial intelligence, and data science. Organized in tree-structure, in descending order of value. Use the Filters on the left to narrow your search. Hover over each button for more help. Best viewed on desktop. Updated 2026-02-23.
Self-taught ML main track - 75% of your learning time.
Machine learning Basics. Let the games begin!
Within deep learning, you'll eventually want to specialize. While exposure to multiple areas is beneficial, focusing on mastering one area initially is often a good strategy. Choose one of the following subfields to dive deeper into.
Focus: Building autonomous AI systems that can reason, plan, use tools, and interact with external services. The hottest applied ML topic of 2024–2026. Covers agent architectures (ReACT, function calling, multi-agent), frameworks (LangChain, LangGraph, AutoGen, smolagents), and protocols (MCP).
Self-taught ML math learning track - 25% of your learning time
Choosing a formal degree path is a significant commitment and replaces the need for a curated self-study path. Choose this if you seek a formal, structured, and credentialed education recognized by employers. This section focuses on the most reputable and cost-effective online options.
Mostly non-technical, supplementary audio for driving / chores / and exercise, or books for leisure learning.
Once you've gotten your fill of the above resources, time to go into auto-pilot. Try a few of these podcasts, find your favorites, and coast.