The AI Engineering Skills Map
GenAI‑101 showed you how AI works. This course teaches you to build and ship real tools with it. Before the details, here is the whole map — so you always know where you are.
Four Top-Level Skills
Andrew Ng breaks AI Engineering into four skills. Three are general; the fourth splits into six of its own.
- Building & deploying AI applications — the six sub-skills (Modules 5–10)
- Software engineering fundamentals — enough to plan well (Module 2)
- Using coding agents — vibe coding as a discipline (Module 3)
- Shaping the build — deciding what to build (Module 4)
The Six Sub-Skills
The heart of the course — the six things it takes to build a reliable AI application.
- LLM foundations · Grounding models with data
- Building agentic systems · Evaluation-driven development
- Operating in production · Machine learning foundations
The Order We Take
General engineering skills come first, so you can plan and direct the build — then the six application sub-skills go deep.
Built From Real Demand
The map was formed by analysing a large number of job postings, structured expert interviews, and survey responses — not invented.
Build It
How to implement: open a notebook and sketch your own version of this map — list the four skills, then honestly rate yourself 1–5 on each. Your lowest scores are where this course will pay off most.
- Weekly AI Tasks tracker — note which map skills your build will lean on: an app (SWE), a coding agent (Module 3), an LLM feature (Modules 5–7).
- Personal brand site — the same, weighted toward SWE, shaping the build, and grounding generation in real material.
What you learned
AI Engineering is four skills; the first splits into six. This course covers the general three first, then goes deep on the six — and each lesson ends by feeding your capstone.