Cindy Zhu.
Free guides
Getting started Multi-tool

The AI learning map: where to learn every piece, free

Hey, it's Cindy 🌱 This is the whole map from the reel. The problem with learning AI isn't a lack of resources, it's not knowing what to learn in what order.
btw the waitlist for AI System Club is now open! If you're keen on using AI to automate your content, video editing, job search, business or build AI agents, feel free to sign up (only if you want to hehe). Ps. the waitlist gets first dibs:)
Read this before you start at step one. You don't have to do all seven, and you definitely don't start at the top if you're already past it. Skim the list, find where you actually are, and go from there. Each one builds on the last, so the order matters more than the speed.
the map

Where to go for each piece, in order

  1. 🌱 Foundations, Elements of AI. Start here if AI still feels fuzzy. The University of Helsinki's free course explains what AI actually is, no code and no math required. Over two million people have taken it. This is your "now I get the big picture" step.
  2. ✍️ Prompting, Learn Prompting. Get genuinely good at talking to AI. This free, open-source guide covers the basics through to advanced moves and prompt security. Prompting well is the single highest-leverage skill on this whole list.
  3. 🧠 How LLMs work, Andrej Karpathy's deep dive. Once prompting clicks, understand what's happening under the hood. Karpathy (co-founder of OpenAI, ex-Tesla) walks through how models like ChatGPT are actually trained in a 3.5-hour deep dive made for normal people. Free on YouTube.
  4. 🤖 Agents, Hugging Face Agents Course. This is where AI stops answering and starts doing. Hugging Face's free course takes you from beginner to building working agents, with hands-on projects and a free certificate. You only need a free account.
  5. 📚 RAG, LangChain's RAG From Scratch. RAG is how you make AI answer from your own documents instead of guessing. LangChain's official playlist teaches it from the ground up in short videos, including the advanced retrieval tricks. Free on YouTube.
  6. ⚙️ Automations, Nate Herk. Now build things that run without you. Nate's channel is the go-to for AI automation and agent workflows, mostly no-code with n8n. Follow along and build a real automation while you watch. Free.
  7. 🔌 MCP, the official Model Context Protocol docs. The deep end, and worth it. MCP is how AI connects to real tools and data. The official tutorial walks you through building your own MCP server step by step. Free docs, and the skill almost nobody has yet.
actually finish it

Turn the map into a plan you'll stick to

A saved list you never open is worthless. Have Claude turn this into a real schedule based on how much time you've got, and quiz you as you go so it sticks.

📆 build my AI study plan 
I want to learn AI from scratch using this exact path, in order: 1) Elements of AI (foundations), 2) Learn Prompting, 3) Andrej Karpathy's LLM deep dive, 4) Hugging Face Agents Course, 5) LangChain RAG From Scratch, 6) Nate Herk for automations, 7) the official MCP docs to build a server.

I can study about [X hours] per week and I'm currently at [total beginner / know some prompting / etc]. Build me a week-by-week plan starting from where I actually am. For each step, tell me what "done" looks like, and give me 3 quick questions at the end of each one to check I actually understood it before I move on.
how to get through it

4 things that make this stick

Don't skip foundations If the later steps feel confusing, it's almost always because a foundation was skipped. Going in order is the whole point of the map.
Build while you learn Don't just watch. Do the prompt, make the agent, wire the automation. The ones you build are the ones you keep.
Use AI to learn AI Paste anything confusing into Claude and ask it to explain like you're new. The tool you're learning is also the best tutor for it.
Depth beats collecting Finishing two of these properly beats bookmarking all seven. Pick where you are and actually complete it.
If you did nothing else this week, do step one or two well. Foundations and prompting carry you further than any advanced topic, and everything after them gets easier once those are solid.
the waitlist is now open

AI System Club is launching soon

Build AI systems with me through recorded lessons you can follow at your own pace. Join the waitlist for first access and the founding price.

btw this list gets first dibs
the moment doors open :)

Privacy · Terms