How to Use YouTube as a Personal Curriculum

A complete computer science curriculum. An MBA's worth of business content. Medical school-level neuroscience explained by a working researcher. All of it is on YouTube, free, from people who are better at explaining their subjects than most paid instructors.

The problem isn't access to learning content. It's building a system that lets you actually use it.

The Structure Problem

Universities provide structure. A schedule, deadlines, a sequence of topics, and accountability. YouTube provides none of that. You can watch a video on linear algebra, then get pulled into a video about historical battles, then watch something about cooking, and feel like you spent time learning without actually building a skill.

Building a learning system on YouTube requires imposing structure from the outside — the algorithm won't do it for you.

Step 1: Define the Skill or Domain

Before you subscribe to anything, decide what you're actually trying to learn. "Technology" is too broad. "How transformer models work" is specific enough to build a curriculum around. "Finance" is too broad. "How to analyze a company's balance sheet" is learnable.

Specific goals lead to specific channel selections. General curiosity leads to an unfocused subscription feed that produces more anxiety than learning.

Step 2: Find the Right Channels for That Domain

For any given topic, there are usually 2-4 YouTube channels that cover it well. For machine learning: 3Blue1Brown (mathematical foundations), Andrej Karpathy (deep dives), and Two Minute Papers (current research). For personal finance: Ben Felix (investing fundamentals), Patrick Boyle (macro), The Plain Bagel (concepts). For web development: Fireship (tools and frameworks), Theo (TypeScript ecosystem).

The goal is depth, not breadth. Two channels that go deep on your topic are more valuable than ten channels that skim the surface.

Step 3: Create a Monitoring System, Not a Watching Queue

The mistake most people make: adding channels to a watch-later list that never gets processed. Videos accumulate, the list becomes an anxiety-producing backlog, and eventually you stop using it.

A better system: monitor channels automatically with audio summaries. When a new video drops from 3Blue1Brown, you get a 4-minute audio summary in Telegram, Discord, or Slack. You listen during your commute. If it's a video you need to watch fully — because the visuals are essential, or because it's on a topic you're actively studying — you add it to your actual watch queue. Everything else, you've absorbed in summary form.

This separates "staying aware of what's being published" from "actively studying a topic." Both are useful. They require different systems.

Step 4: Active Review for Deep Learning

Passive listening builds awareness. Active review builds skill. For the videos you watch fully, take notes in your own words. Explain the concept back to yourself. Connect it to something you already know. This is basic spaced repetition applied to video content.

Notion, Obsidian, or even a simple notes file works here. The format doesn't matter. The act of converting what you watched into your own words matters.

The Sustainable Version

A sustainable YouTube learning system looks like this: you follow 5-10 channels in your specific domain. Every new video gets summarized and delivered to you automatically. You listen to summaries during commutes and breaks. Twice a week, you watch one full video from your "worth watching" queue. You take notes on what you watch fully.

That's a few minutes of passive learning every day and 30-40 minutes of active study per week. Consistent over months, it adds up to genuine knowledge — built from some of the best educational content ever created, without the chaos of an unstructured YouTube feed.