Our project attempts to package a generative AI tool in a preconfigured development environment, with all the tools needed for a student to start programming. The generative AI chatbot is configured by instructors and is designed to guide students through problems they encounter when programming, without spoon-feeding them the solution. The chatbot is able to provide a personalized learning experience for students, without burdening instructors. Common questions the chatbot receives are collected and forwarded to instructors to help them identify problems that students face.

At HKUST, C++ courses are a mandatory course for students majoring in computer science and other related majors. And yet, many students struggle in the course every year, and find it difficult to keep up with the large amount of content covered in the curriculum.

As generative AI improves and becomes more widespread, we believe that it can be a valuable resource in aiding these students that are struggling. By leveraging the capabilities of generative AI, we would like to provide students with personalized assistants that can teach and guide them through problems they encounter when learning the course material. By doing so, we are able to address and tackle students’ issues individually, providing targeted help to those who are struggling. Considering the large scale nature of the course, such “personalized” tutoring is not feasible for instructors to provide themselves.

Our solution provides a simple but intuitive way to leverage the ability of gen AI while not undermining the learning objectives of the course, helping students grasp and learn fundamental knowledge needed in this field. This will not just be beneficial to HKUST students but can also be extended to other schools and courses.

N/A

| Function Point | Technical Application (specific technical points) | Progress |
|-----------------------------------------------------|-------------------------------------------------------------------------------|------------------------------------------------------------------------------------------------|
|GitHub Codespace Configuration | Docker | Done |
|VSCode extension development | TypeScript, VSCode Extension API, GitHub Copilot API | Final modifications in progress |
|Backend development | Python, FastAPI, Azure OpenAI API | Updating along with addition of frontend and extension features |
|Database configuration | MySQL | Done |
|Frontend admin portal development | TypeScript, ReactJS | Feature building in progress |

This project is the pilot study to use gen ai tool to facilitate and assists students and teachers in coding education

Our project reimagines generative AI as a course goal aligned mentor rather than a generic assistant in following aspects:
- Course curriculum grounded tutoring with specific reference to lecture note page.
Instead of utilizing the chatbot as a tool that works for you, we configured the chatbot to act as a mentor and teacher that walks through the problem with the student, guiding them through the problem solving process and reinforcing their fundamental understanding of the subject.

- Teacher-student interaction and understanding through gen AI usage analytics.
By analyzing the questions posed to the chatbot, we are able to determine what topics and concepts students struggle with most, and forward this feedback towards instructors. Instructors can use this data to reflect on the structure of the course and their teaching, refining and improving the course to address areas that pose a challenge to students.

-Personalized exercise generating with test cases.
They say practice makes perfect, and it may be difficult for students to find exercises and questions that allow them to reinforce their understanding of the course content. To help students in this regard, our tool is capable of generating exercises of varying degrees of difficulty for students to evaluate themselves and check their understanding, and grade their work afterwards.

Since the project assists instructors in providing aid to students, it is easily scalable to target a large number of students. One on one tutoring and grading capabilities are all done through generative AI, allowing help to be provided to a large number of students without greatly increasing the workload on instructors and teaching staff.

This project tackles overreliance issue of students on AI that may hinder their learning especially in University setting where self-motivated learning is necessary.

While generative AI has a lot of potential, it is crucial that it is not used as a substitute for critical thinking. Many times have people blindly trusted the content generated by generative AI, and we condemn the use of generative AI in cheating or dishonest academic behavior. However, outright banning the use of generative AI is impractical and not a feasible solution to this problem. We hope to bridge the gap by encouraging the use of generative AI as a tutor, an assistant to guide and stimulate thought in students. To accomplish this, we enforce various restrictions towards the chatbot and encourage it to explain and teach, as opposed to doing the work for students.

With the help of our project, everyone will be able to learn how to code regardless of their background, and master the fundamental concepts of programming with the assistance of generative AI.

Personal Information Collection Statement (PICS):
1. The personal data collected in this form will be used for activity-organizing, record keeping and reporting only. The collected personal data will be purged within 6 years after the event.
2. Please note that it is obligatory to provide the personal data required.
3. Your personal data collected will be kept by the LTTC and will not be transferred to outside parties.
4. You have the right to request access to and correction of information held by us about you. If you wish to access or correct your personal data, please contact our staff at lttc@eduhk.hk.
5. The University’s Privacy Policy Statement can be access at https://www.eduhk.hk/en/privacy-policy.

• I have read and agree to the competition rules and privacy policy.