In this project you will design a complete e-learning course outline using prompts — learning outcomes, modules, lessons, assessments, and a launch-ready syllabus. The deliverable is a Markdown document a course platform like Teachable, Thinkific, or your own LMS could ingest directly.
Most online courses are written from the inside out: someone records a few lessons, slaps a title on top, and hopes a structure emerges. The result is a course full of "interesting content" that students never finish. A good course is designed from the outside in: outcomes first, then the smallest set of modules that delivers those outcomes, then lessons inside each module, then assessments that prove the outcomes were reached.
This project shows that outside-in design as a prompt chain. We will build the course for a fictional topic: "Confident SQL for Analysts", a four-week beginner-to-intermediate course.
The chain mirrors instructional design's "backward design" model. Five focused prompts feed into one syllabus.
Forward design
Give me an outline for a SQL course.
You get a chapter list: "SELECT, WHERE, JOIN, GROUP BY, subqueries, window functions". It looks like a textbook table of contents and treats topics as the goal. Students complete every module and still cannot answer the only question that matters: "Can you now write a query to answer a real business question?"
Step 1 — Outcomes prompt
You are an experienced instructional designer.
Course title: "Confident SQL for Analysts"
Audience: business analysts and product managers with zero SQL
experience but comfort using Excel formulas.
Duration: 4 weeks, ~3 hours per week.
Constraint: by the end, learners should be able to answer real
business questions using a real database, not memorise syntax.
Write 6 learning outcomes for this course.
Rules:
- Each outcome starts with a verb (Bloom's: apply, analyse,
evaluate, create — not "understand" or "know").
- Each outcome is concrete enough to test.
- No two outcomes overlap.
- Cover from beginner to intermediate; do not include advanced
topics like CTEs unless they support a beginner outcome.
Sample outcomes: "O1: Write SELECT queries that filter, sort, and limit rows to answer a specific business question. O2: Combine data from two related tables using INNER and LEFT JOIN. O3: Aggregate rows with GROUP BY to compute KPIs (totals, averages, ratios). O4: Diagnose why a query returns the wrong number of rows…"
Step 2 — Modules prompt
Map the 6 outcomes above into 4 modules (one per week).
For each module:
- module title (states the outcome in plain English, not jargon)
- which outcome(s) this module is responsible for
- one "real business question" the learner will be able to answer
by the end of the module
- 2–3 misconceptions the module must address
- a rough total minutes count (target ~3 hours per module)
Return as a Markdown table.
Step 3 — Lessons prompt
Expand Module 2 ("Joining tables to combine data") into 5 lessons.
For each lesson return:
- lesson title (action-oriented, not topic-oriented)
- estimated runtime (8–15 min)
- type: video / reading / interactive / quiz
- learning objective (one sentence, starts with a verb)
- key concepts (≤3 bullets)
- example to walk through (a real-sounding business scenario)
- common student mistake to address explicitly
- transition: how this lesson sets up the next
Voice for lesson titles: friendly and direct.
Forbidden words in titles: "Introduction to", "Understanding",
"Basics of".
Output as Markdown.
Sample lesson: "Lesson 2.3 — When two LEFT JOINs disagree. 12 min · video. Objective: diagnose a query whose row count doubles after a second join. Common mistake: assuming joins multiply rows independently…" Notice how this title reads — it is a problem to solve, not a topic to memorise.
Step 4 — Assessments prompt
Design assessments that prove the 6 outcomes were achieved.
For each outcome:
- one formative quiz question (multiple choice or short answer)
used inside a lesson
- one summative task (a small SQL exercise on a real-feeling
dataset) used at the end of the relevant module
- one capstone task at the end of the course that combines this
outcome with at least one other
Rules:
- Tasks must require applying the skill, not recalling syntax.
- Provide one realistic sample dataset description per task
(no actual data needed yet).
- For each task, write a 3-line marking rubric (excellent / okay /
not yet) so a TA could grade consistently.
Step 5 — Syllabus prompt
Compose the final course syllabus as a single Markdown document.
Order:
1) Course title + one-paragraph promise (what the learner will be
able to do after 4 weeks)
2) Who this course is for / not for
3) Pre-requisites (be honest)
4) The 6 learning outcomes
5) Week-by-week breakdown using the module + lesson data
6) Assessments + final capstone description
7) Tools and access required (e.g. "a free PostgreSQL playground")
8) Estimated total time commitment
Voice: warm, practical, slight British understatement.
Forbidden: "transform your career", "in just 4 weeks", "industry-
leading", any hype. Use plain language.
The output is a launch-ready syllabus you could paste into a course platform, send to a co-instructor, or publish as the course landing page's "What you will learn" section. Save it as course-syllabus.md.
Tip: After step 3, sanity-check the lesson count against the time budget. A 4-week course at 3 hours per week is roughly 12 hours; if your lessons add up to 18 hours, cut something now — not after recording starts.
Pick a topic you know well enough to teach. Run the outcomes prompt. Count how many outcomes start with "understand" or "know". Force yourself to rewrite each one using a Bloom verb. The course changes shape with the verbs.
Take your modules and run a "real business question" test: for each module, write the question the learner will be able to answer by the end. If you cannot write that question, the module is not focused enough — redesign it.
Generate a capstone-only prompt for an existing course you have taken. What single project would prove the course's outcomes were achieved? Often the answer reveals exactly what was missing from the course.
Sign in to join the discussion and post comments.
Sign inAdvanced Prompt Engineering Techniques
Master the powerful techniques AI experts use every day. Chain-of-thought, RAG, agents, function calling, prompt evaluation, and much more — 20 deep-dive tutorials.
Prompt Engineering for Specific AI Tools
Tool-by-tool mastery — deep dives into ChatGPT, Claude, Gemini, GitHub Copilot, Midjourney, Stable Diffusion, and more. Learn the exact prompting techniques each platform rewards.
Prompt Engineering for Content & Copywriting
Write blogs, ads, emails, and social media content ten times faster with AI. 13 practical tutorials on prompt engineering for content creators and copywriters.
Prompt Engineering for Data Science & Analytics
Supercharge your data workflows with AI. 15 practical tutorials on using prompt engineering for data cleaning, EDA, machine learning, SQL, visualisation, and more.
Foundations of Prompt Engineering
The must-know basics of prompt engineering. Learn what prompts are, how AI models read them, and how to write clear instructions that get great results.
Prompt Engineering for Business & Productivity
Use AI to work smarter — automate tasks, make better decisions, and communicate professionally. 12 practical business prompt tutorials for professionals.