Personal Experience with AI
10 Dec 2024
My use of Artificial Intelligence in ICS 314
I. Introduction
The integration of Artificial Intelligence (AI) in education, especially in fields like Software Engineering, marks a revolutionary shift in how we learn and apply complex concepts. In ICS 314, AI tools such as ChatGPT, Bard, and Co-Pilot have emerged as pivotal resources. These tools not only assist in understanding theoretical aspects but also in applying software engineering principles practically. My experience with AI in this course has been a journey of exploration, understanding its potential and limitations in a learning environment.
II. Personal Experience with AI
1. Experience WODs
For WODs, I use ChatGPT most of the times, found that the AI was still able to help me conceptualize and complete the WODs quite well under normal circumstances
2. In-class Practice WODs
For in-class Practice WODs, I usually don’t use AI because the class is still focused on learning to learn to practice how to complete the WODs.
3. In-class WODs
In class WODs,I usually use AI, the first point is because AI really saves me a lot of time and usually AI generates some content to help you understand.
4. Essays
AI played a significant role in essay writing. I used ChatGPT to structure my essays, especially for technical explanations. The prompt typically was, “Outline an essay explaining [software engineering concept].” This provided a good starting point but needed further personal input for depth and accuracy.
5. Final Project
For the final project, AI was instrumental in initial brainstorming and concept validation. I asked ChatGPT, “What are innovative project ideas involving [specific software engineering topic]?” The ideas were creative but needed feasibility checks.
6. Learning a Concept/Tutorial
AI tools were helpful for understanding new concepts. For instance, when learning about API integration, I used the prompt, “Explain API integration in TypeScript with examples.” The explanation was basic but a good starting point.
7. Answering a Question in Class or in Discord
I occasionally used AI to formulate responses to questions asked in class or on Discord. For complex queries, AI provided a foundational understanding which I then expanded upon.
8. Asking or Answering a Smart-Question
In these instances, AI was less utilized, as the focus was more on personal interpretation and problem-solving skills.
9. Coding Example
For coding examples like using Underscore’s .pluck, I used the prompt, “Provide a TypeScript coding example using Underscore’s .pluck.” The result was a straightforward code snippet, useful for quick understanding.
10. Explaining Code
I found AI less effective for explaining specific code segments due to the lack of context understanding.
11. Writing Code
For writing code, AI tools like Co-Pilot were valuable. They offered code suggestions that often needed adjustments to fit the exact requirements.
12. Documenting Code
AI was not used extensively for code documentation, as I preferred to write these manually for better understanding and clarity.
13. Quality Assurance
I tried using AI for quality assurance with prompts like, “What’s wrong with this code [code snippet]?” The results were hit-or-miss, often requiring further manual inspection.
14. Other uses in ICS 314
In other aspects like project management and team collaboration, AI’s role was minimal, as these areas required more human-centric skills.
III. Impact on Learning and Understanding
The incorporation of AI in ICS 314 has been a double-edged sword. On one hand, it has enhanced comprehension and provided quick access to information and coding assistance. On the other, there’s a risk of over-reliance which can impede the development of independent problem-solving skills. AI has both clarified and, at times, oversimplified complex software engineering concepts.
IV. Practical Applications
Outside ICS 314, I’ve observed AI’s application in real-world projects like those in hackathons. AI tools have been effective in providing coding suggestions, debugging assistance, and even in generating initial project ideas. However, their effectiveness is contingent on the complexity of the task and the user’s ability to integrate AI insights effectively.
V. Challenges and Opportunities
A major challenge has been the AI’s limited ability to comprehend context and nuanced software engineering problems. However, there’s significant opportunity in further integrating AI into more practical, hands-on aspects of software engineering education, like live coding sessions or project management simulations.
VI. Comparative Analysis
Comparing traditional teaching methods with AI-enhanced approaches reveals a significant difference in engagement and accessibility to information. While traditional methods offer structured learning, AI provides flexibility and immediate assistance. However, knowledge retention and practical skill development still heavily rely on traditional methods, as they offer deeper, more contextual learning experiences.
VII. Future Considerations
Looking ahead, AI’s role in software engineering education is poised to grow. Advancements