Vamsi

Professor's View: Seeing a summary of what students are confused on
Student's View: Options they can choose using the Vamsi VSCode Extension

Inspiration

With larger classes, it can be hard for professors to understand all the issues that students have. We wanted something that could help students code and help professors identify weak points for students, so professors know what concepts to focus on in class. We got inspired by a mentor we once had, who was very attentive and great and supplemented our learning needs. We want that to be available to every student to further their studies.

What it does

Vamsi has two parts. The first part is a VS Code extension that utilizes Gemini AI to give audible tips on how to solve programming problems given by the professor. It reads the contents of the student's problem file to provide relevant hints in small working sessions during the lecture. The second part is a website that the professor uses to control the questions as well as get a summary of the issues that students faced, identifying concepts that may require more explanation.

How we built it

We used the VS Code Extension template, Gemini API and ElevenLabs text-to-speech to create the VS Code extension. We used Node.js and Prisma to create the front and backend of the professor's website.

Challenges we ran into

We had wanted to use the "Professional Voice Clone" from ElevenLabs to create a custom voice, however, this is a paid feature not included in our free plan.

We were too good at cup stacking and won too hard

Accomplishments that we're proud of

We're very proud to have created an AI that can help students and professors. We were able to learn new things such as how to create a VS code extension and use Gemini API, things that none of us had ever done before.

What we learned

We learned how to use Gemini API, set up a VS Code extension, use ElevenLabs products and more.

What's next for Vamsi

At the moment, we're using our student plans for Gemini API and the free trial for ElevenLabs. Obviously, if this were to be used at scale, we would run out of credits very fast. We would need to get business plans to ensure we won't run out of credits. To avoid this issue, we may allow users to use their own API keys.

We would also like to do field tests with actual students and a professor during class, so we can see how Vamsi would work in an actual class environment. We would take feedback from this test to improve Vamsi.