Coding with AI

Course Overview

Students are introduced to the main goal of the course.

Rather than focusing solely on writing code manually, students learn how to build software effectively with the assistance of modern AI tools while staying in control of the development process. oaicite:2

Software Development is Problem Solving

Programming does not begin with typing code.

Students learn that the developer's primary responsibility is to:

• Understand the problem.
• Understand the user.
• Break large problems into smaller pieces.
• Design the solution before implementation.

AI assists with writing code, but humans solve the problem. oaicite:3

Where Every Project Begins

Before opening a code editor or asking AI to generate code, students answer three questions:

• Who is the application for?
• What problem does it solve?
• What is the smallest useful version that can be built?

These questions form the foundation of every successful software project. oaicite:4

The Software Development Lifecycle

Students are introduced to the seven stages of software development:

1. Idea
2. Requirements
3. Design
4. Build
5. Test
6. Ship
7. Maintain

The lesson emphasizes that most software work happens before writing code. oaicite:5

Anatomy of an Application

Every application consists of three major components:

• Frontend – what users see.
• Backend – the application logic.
• Data – persistent information stored in a database.

Students use these concepts throughout the semester when discussing application architecture. oaicite:6

Understanding AI Coding Assistants

Students learn how AI actually works.

Key ideas include:

• AI predicts likely code rather than understanding the project.
• AI can produce incorrect answers confidently.
• AI only knows the information provided in the prompt.

These limitations explain why developers must review AI-generated work carefully. oaicite:7

The Core Loop

The course follows a consistent AI-assisted workflow:

1. Specify
2. Plan
3. Review
4. Build
5. Verify
6. Iterate

Students practice reviewing AI output before accepting it, ensuring they remain responsible for every decision. oaicite:8

Good AI Habits

Students compare productive AI usage with poor AI usage.

Good habits include:

• Asking for plans before code.
• Requesting explanations.
• Reviewing generated code.
• Using AI as a tutor.

Poor habits include:

• Copying code without understanding it.
• Letting AI make design decisions.
• Asking AI to build an entire application in one prompt.
• Shipping code without verification. oaicite:9

The One Rule

Throughout the semester, one rule applies:

You may only submit code you can explain.

Students are encouraged to ask AI for explanations until they fully understand how the code works. oaicite:10

Course Deliverables

Students work in groups to produce:

• A small but functional application.
• A reflection log documenting how AI was used during development.

The reflection log carries significant weight because the course emphasizes understanding and decision-making rather than simply producing code. oaicite:11

Development Tools

Students are introduced to the tools used throughout the course:

• Claude or ChatGPT
• Cursor or Windsurf
• GitHub Copilot
• GitHub

Each tool has a specific role in the AI-assisted development workflow. oaicite:12

Week 1 Tasks

Before the next class meeting, students should:

1. Set up all required development tools.
2. Create their first reflection log entry.
3. Write a short reflection on what "coding with AI" means to them.

These activities prepare students for the hands-on lessons that follow. oaicite:13