Why should you care?

Every modern product team touches AI somehow. If you cannot define AI in plain English, you will struggle to choose the right tool later (rules vs ML vs LLM).

See it live — copy this example

Treat examples as Python-style notes you can paste into a notebook or rewrite in your stack. Prefer public sample data — never real private records.

# Think in problems, not buzzwords
problem = "Suggest next product for a shopper"
approach = "recommendation model or simple rules + popularity"
print(problem, "→", approach)

What happened?

• Good AI work starts by naming the user problem and a possible approach.
• Not every problem needs a neural network.

Practice next

1. Write three AI problems from apps you use daily.
2. For each, say: rules, classical ML, or generative AI.
3. Create a folder named your AI project for notes.
4. Add a fourth problem from your job or college.
5. Mark which ones need private data.

Remember

AI solves judgment-like tasks with data/models. Start from the user problem. your AI project is this course’s project theme.

Product triage

A startup wants “AI” on the roadmap.

Outcome: You force a clear problem statement before picking a model.