Learning · Agentic Operations
Agentic Operations
Learn to build AI agents, one real concept at a time.
This course is four levels, each building on the last. You move on when you can explain the idea yourself and the hands-on exercise works — not just when you've read the page. Keep your own AI assistant (Claude, ChatGPT, whatever you use) open alongside these lessons; working through follow-up questions with it is part of the method, not a detour.
Start here — the mental-model foundation
These four lessons are the foundation of the whole course — one per level, best read in order. Everything else below is hands-on practice built on top of them.
- ✓0001 — Not in Context = Doesn't Exist (Level 1)
- ✓0002 — Match the Pattern to the Problem (Level 2)
- ✓0003 — Traces Before Prompts (Level 3)
- ✓0004 — Own the Feedback Loop (Level 4)
The practice ladder
Each level's hands-on lessons build on its foundational lesson. Work them in order; each level ends with a gate you pass by demonstrating the skill, not by reading about it.
Level 1 · Foundations
Level 2 · Developing
- ✓0007 — Structured Prompting & the Injection Lab
- ✓0008 — The ReAct Multi-Hop Build
- ✓0009 — RAG Build & Break
- ✓0010 — Episodic Memory (gate)
Level 3 · Advanced
- ✓0011 — The Orchestrator Bake-Off
- ✓0012 — Tracing & the Planted Regression
- ✓0013 — The Critic/Verifier Step (gate)