Design Philosophy
- Teach concepts before tools
- Build before discussing trends
- Practice teaching every week
- Assess understanding through demonstration, not quizzes
PHASE 1 — Cognitive Foundations (Weeks 1–4)
Goal: True conceptual understanding of AI mechanics
Module 1: What Intelligence Actually Means
Focus
- Human vs machine intelligence
- Pattern recognition vs reasoning
- Automation vs AI
Trainer Skill
Explain AI to a 10-year-old and a CEO using different analogies.
Deliverable
Create a 3-minute teaching mini-lesson.
Module 2: Models and Representation
Focus
- What a model is (not buzzwords)
- Inputs → transformation → outputs
- Why data must be encoded numerically
Trainer Skill
Use real-world analogies for models.
Deliverable
Whiteboard explanation of how a spam filter works.
Module 3: Data Is the Real Engine
Focus
- Structured vs unstructured data
- Bias sources
- Garbage-in-garbage-out
Trainer Skill
Help learners diagnose bad datasets.
Deliverable
Analyze a flawed dataset and explain its problems.
Module 4: Training vs Using a Model
Focus
- Training phase
- Inference phase
- Overfitting
- Underfitting
Trainer Skill
Explain training without math.
Deliverable
Teach the concept using only physical objects.
PHASE 2 — Core Machine Learning Literacy (Weeks 5–8)
Goal: Understand what models actually do under the hood
Module 5: Supervised Learning
- Classification
- Regression
- Labels
- Decision boundaries
Build: simple classifier with spreadsheet logic
Module 6: Unsupervised Learning
- Clustering
- Similarity
- Feature space
Build: manual clustering exercise
Week 7: Reinforcement Learning
- Rewards
- Exploration vs exploitation
- Policy learning
Simulation: decision-reward game
Module 8: Model Evaluation
- Accuracy vs usefulness
- Precision/Recall
- Real-world failure cases
Deliverable: critique a real AI system failure.
PHASE 3 — Building Real AI (Weeks 9–12)
Goal: Hands-on creation + debugging ability
Module 9: Build Your First Model
Tool: simple Python notebook or visual builder
Students must:
- load data
- train model
- test output
Module 10: NLP System
Build:
- text classifier or chatbot logic
Focus:
understanding pipeline, not tool buttons.
Module 11: Vision System
Build:
- simple image classifier
Focus:
interpret model mistakes.
Module 12: Debugging Models
Most courses skip this. Trainers cannot.
Learn:
- diagnosing bad predictions
- fixing training problems
- improving datasets
Final build:
Fix a broken model.
PHASE 4 — Teaching AI Mastery (Weeks 13–16)
Goal: Produce confident instructors
Module 13: Pedagogy for Technical Topics
Learn how people misunderstand:
- probability
- randomness
- statistics
- algorithms
Practice:
explain concepts without jargon.
Module 14: Lesson Design
Create:
- objectives
- exercises
- examples
- assessments
Deliverable:
Full lesson plan.
Module 15: Teaching Practicum
Students teach a real lesson.
Peers evaluate:
- clarity
- pacing
- accuracy
- engagement
Module 16: Capstone
Final demonstration:
Teach a complete AI topic from scratch
to beginners who know nothing.
Must include:
- explanation
- analogy
- example
- activity
- assessment