Learn prompt engineering for real work

Overview

Every PM, marketer, designer, and operator now writes prompts. Most of them are bad, long, vague, drifting, irreproducible. This course gives you the small set of patterns that turn prompts from craft-by-vibes into a real, transferable skill: structure, decomposition, role assignment, and the iteration loop that gets you from 'kind of works' to 'works every time'.

What you'll learn

By the end, you'll be able to do these, not just have read about them.

• 01

  Move from intuition to a repeatable prompt-design process

• 02

  Apply structure, examples, and constraints that hold up across inputs

• 03

  Iterate prompts with a small eval loop instead of vibes

• 04

  Recognize when a prompt cannot fix what only fine-tuning can

Suggested chapters

This is the typical chapter list. Your version is generated against your background and adapts as you go. It may compress, expand, or reorder these.

1. 01

   Why prompts work (and don't)

   A grounded mental model, context, attention, instruction-following, without the jargon.

2. 02

   Structure over cleverness

   The 5-part prompt skeleton, role assignment, and how to reduce drift.

3. 03

   Decomposition

   Breaking big tasks into chained, checkable steps, your single biggest unlock.

4. 04

   Examples that pull weight

   Few-shot done right, when it backfires, and how to swap for instruction.

5. 05

   Iteration & evaluation

   Tight feedback loops, golden cases, regression checking, for non-engineers.

6. 06

   Prompts in real workflows

   Email, briefs, analysis, creative, patterns by job-shape, not by tool.

7. 07

   Building your prompt library

   Versioned, parameterized prompts you actually reuse, and improve over time.

Common questions

• Is this technical?

  No. There's no code. Engineers welcome, but the course is built for any operator who uses LLMs day-to-day.

• Will it teach me to build agents or RAG?

  No, those are separate courses (AI Agents, RAG From Scratch). This is about the foundational craft of prompting.

• Is this a fixed course, or is it built for me?

  Built for you. The chapter list below is a typical outline. Your actual course is generated against your role, experience, and what you already know, then adapts as you go.

• How long does it take?

  Most learners finish in 2–6 weeks at a normal pace, depending on the topic. Octo compresses where you're strong and slows down where you're weak.

• Is there a fixed schedule or cohort?

  No. You start when you start. There's no live session, no calendar, no deadline.

• Can I ask questions while I'm learning?

  Yes, every module has an AI Sidekick in the margin. Ask for a different example, push back, or get a clarifying analogy without leaving the page.

• What do I get at the end?

  A verifiable, HMAC-signed certificate with a public verify page. It records the modules passed, scores, and capstone, not just attendance.

• How much does it cost?

  Octo is in research preview, courses are open. We'll be transparent before pricing changes.