Assignment: AI pair programming with hardware verification

1. Objective

The goal of this assignment is to enhance your programming and engineering skills by using an AI tool as a pair programmer. You will implement a project of your choice from the provided list. A key requirement is that you must verify your final code using a physical hardware setup. Throughout this assignment, you will interact with an AI model to help you develop the project, while carefully documenting your prompts, errors, and thought processes in a reflection journal.

2. Project guidelines

All students must work on an independent Python project with a clear hardware validation component or a prototype that informs the data for the project. This project must demonstrate the following five key elements of Python programming that you learn in this class. An ideal project will:

1. Control the program flow with branching and loops.
2. Use correct data structures for optimal computations, as relevant for engineering.
3. Use functions for modular code.
4. Load data from real circuits/hardware stored as files (or data available on the internet), write outputs to files (if needed), and visualize outputs in graphs (as applicable).
5. Document the flow of logic with comments, docstrings, and user-friendly messages.

What project can I work on?

A few example project ideas are provided in the project_ideas.md document, but you are more than welcome (in fact, encouraged!) to come up with your own ideas that satisfy the project requirements.

3. Introduction to AI pair programming

What is Pair Programming? Pair programming is a software development technique where two programmers work together. One, the "driver," writes code, while the other, the "navigator," reviews each line of code as it's written. This collaborative approach helps identify errors quickly and promotes knowledge sharing. In this assignment, your pair programmer is an AI tool. You will work with it back-and-forth to develop code, identify errors, and complete your project.

What are Generative AI Tools? You have likely heard of Generative AI tools like ChatGPT, Google Gemini, or Microsoft Copilot. These tools can generate human-like text, including code, based on input prompts. They can assist in programming by providing suggestions, generating code snippets, and offering explanations.

Important Note: While an AI can be a powerful partner, it is not a replacement for your understanding. Use the AI to get ideas or initial code, but you must ensure you understand the code fully. Your role is to control the creative and critical aspects of your project and to build the physical hardware, which the AI cannot do.

4. Assignment tasks

Step 1: Project selection

Choose a project from the project_ideas.md document. Select one that you find interesting and that you believe is achievable for your skill level.

Step 2: Start a reflection journal

You must keep a reflection journal throughout the project. This will be a key part of your submission. Document your thought process, project updates, struggles, and what you learn along the way.

Step 3: Initial interaction with the AI

Begin by signing up for an AI tool (like ChatGPT or Google Gemini) and starting a new conversation. Your first prompt should set the context for the AI.

Starter prompt example: Hello! I'm a student learning Python. I need to build a project to create a [your chosen project idea, e.g., 'temperature logger']. I'd like you to act as my pair programmer. Please guide me through setting up the initial framework for this project, providing explanations along the way. Do you understand? Please say "yes" or "no" and do not write any code yet. Record this first prompt in your journal.

Step 4: Iterative development & prompt engineering

For each feature you want to add to your project: 1. Engineer a Prompt: Write a clear prompt to the AI asking for help. For example: "I'd like to add a feature to log the temperature to a CSV file every 10 seconds. Can you show me how to implement this?" 2. Implement and Test: Incorporate the AI's suggestions into your code and test it thoroughly to ensure it works.

Step 5: Error handling

When you encounter errors, try to debug them on your own first. Use print() statements or your debugger to understand the problem. If you are still stuck, ask the AI for help.

Example error prompt: I'm getting a `TypeError` when I run my code. The error is: 'unsupported operand type(s) for +: 'int' and 'str''. Here is my code: [paste your code snippet]. Can you suggest what I should check? Document the errors you encounter and how you solve them in your journal.

Step 6: Hardware verification

This is the most important step. You must build the physical circuit for your project and verify that your code works with the hardware. For example, if you built a temperature logger, you must show the sensor collecting real data. If you built a traffic light, you must show the LEDs lighting up in the correct sequence.

Step 7: Finalizing the project

• Review your entire codebase and make sure you understand every part.
• Add comments to your code to explain what it does.
• Confirm that the project runs smoothly without errors.

5. Submission requirements

You will submit the following items, typically in a single .zip file:

1. All code files: All .py, .ino, or other source code files for your project.
2. Hardware verification video: A short video (1-2 minutes) clearly demonstrating that your project's hardware is working as controlled by your software.
3. Reflection journal (PDF): Your journal should include:
   • An introduction describing the project and its objectives.
   • A log of your thought process and development journey.
   • A clean list of the final, successful prompts you used with the AI.
   • A reflective essay on what you learned, what was challenging, and your thoughts on using AI for programming.
4. AI interaction transcript (PDF): The full, unedited conversation history with your AI partner from the beginning to the end of the project.