AI Literacy: From Prompts to Agent Teams
A three-module course that takes you from writing better prompts to running multi-agent workflows — with a short video and hands-on exercise for every lesson.
No coding required. Built for anyone who works with AI and wants to stop guessing and start getting results.
Why AI literacy matters
Knowing how to use AI effectively is becoming as fundamental as knowing how to use email or a spreadsheet. But unlike those tools, AI has a learning curve that isn't obvious — because it responds to language, it feels like you should already know how to use it.
The gap between a mediocre prompt and a great one isn't talent. It's understanding a few key principles about how these systems work. That's what this course teaches.
By the end, you'll be able to:
- Get dramatically better results from any AI tool, immediately
- Build a personal knowledge system that makes your AI interactions smarter over time
- Coordinate multiple AI agents on complex tasks the way a manager coordinates a team
Course Modules
What each module covers
Module 1: Reverse Prompting
Most people approach AI by trying to craft the perfect question. The better approach: let the AI ask you the questions.
Reverse prompting means starting with a goal and asking the AI to surface what it needs to know before it begins. Instead of spending 20 minutes writing a prompt, you spend 2 minutes describing your goal and let the model identify the gaps.
This module teaches you:
- Why most prompts fail (and how to diagnose yours)
- The reverse prompting framework: goal, context, constraints, format
- How to turn a vague request into a precise brief — by asking the AI to help you write it
- How to recognize when a prompt is ready versus when it needs more work
Example lesson: You want to write a performance review for a direct report. Instead of starting with "write a performance review," you start with: "I need to write a performance review. Before you draft anything, ask me the questions you need answered to do this well." The model interviews you — and the result is far better than anything you'd have written directly.
Module 2: Memory and Context
AI models don't remember you. Every conversation starts fresh. This is one of the biggest limitations in practical AI use — and it's completely solvable once you understand it.
This module teaches you to build a context layer: a personal system of documents, templates, and background information you bring into your conversations. With the right setup, every AI session picks up exactly where you left off.
This module teaches you:
- How context windows work and why they matter
- Building a personal "AI briefing document" that you paste into every session
- Project memory: how to maintain continuity across a long-running task
- When to use AI Projects and persistent memory features versus doing it manually
- How to compress context so you can fit more into a conversation
Example lesson: You're working with a client over several months. Each time you open a new AI session, you paste a one-page client brief: their industry, key contacts, recent discussions, and open questions. The AI immediately has everything it needs — no re-explaining required.
Module 3: Multi-Agent and Delegation
The biggest productivity gains from AI come not from using one model for one task, but from orchestrating multiple specialized agents working in parallel.
Think of it like building a team. A good manager doesn't do everything themselves — they decompose work into roles, assign each to the right person, and synthesize the results. You can do the same with AI agents.
This module teaches you:
- How to decompose a complex task into parallel workstreams
- Assigning roles: researcher, writer, reviewer, critic, synthesizer
- Running agents in sequence (pipeline) versus in parallel (team)
- How to handle disagreements between agents — and why that's often a feature
- When to bring a human back into the loop
Example lesson: You need to write a comprehensive market analysis. Instead of asking one AI to do everything (and getting a surface-level result), you run three agents in parallel: one researches the competitive landscape, one analyzes recent news, one reviews your existing internal data. A fourth agent synthesizes their outputs into a coherent report. The result is deeper and faster than any single agent could produce.
How to use this course
Each lesson has three parts:
- Watch — a short video framing the idea (most under 2 minutes)
- Read — the full explanation with examples, patterns, and context
- Practice — a hands-on exercise using your own real work
Work through lessons in order the first time. The concepts build on each other — Module 1 is the foundation for everything in Modules 2 and 3. Once you've completed the course, you can return to individual lessons as reference.
The exercises use real work, not toy examples. You'll leave each lesson with something actually useful — a better prompt for a task you do regularly, a context document for a current project, or a delegation pattern for something on your plate right now.
Time commitment: Each module takes about an hour. You can do one module a week or work through the whole course in an afternoon. The exercises take as long as the work itself takes — which is the point.
Who this course is for
This course is for people who use AI tools at work and want to get dramatically more out of them. You don't need a technical background. You need to be willing to try things and pay attention to what works.
If you've used ChatGPT, Claude, or similar tools and felt like you were getting mediocre results — this course is for you. The gap between average and excellent AI use is mostly technique, not access.
If you've never used an AI tool before, start with the Quick Start guide first, then come back here.