Inspiration

Most people are nervous before a job interview, so we wanted to create an application to help change that.

What it does

Our Unity environment will simulate participating in an actual interview with a potential employer in VR. ChatGPT will generate interview questions based on a form that the user fills out before initiating the simulation. After the user answers the question, chatGPT will provide feedback on their answer.

How we built it

First, we created a 3D, virtual environment in the Unity engine optimized for VR on Meta's Oculus headset. Once the environment was completed, we used a C# script to integrate chatGPT and OpenAi's API key into Unity. The finished product generates both questions and reponses for the virtual mock interview.

Challenges we ran into

We had trouble getting the C#script to work, as it was our first time as a team working with the language as in depth as we had done for the project. Additionally, we had an issue submitting files to GitHub, as our final file size for the Unity Environment was quite large.

Accomplishments that we're proud of

Despite the challenges we faced, we were able to create a working script that integrated chatGPT into Unity successfully, and created a video demonstration that shows this success in action. Also, we were able to solve our uploading issues and were able to post all of our files to GitHub.

What we learned

We all learned a lot about C#, as well as how to use .gitignore files to properly submit large Unity files to a GitHub repository. We even learned how incorporate a ChatGPT API key into Unity without having to pay for a subscription fee, and still have a working AI avatar in Unity for the user to interact with upon doing so. Finally, we learned how to initialize this environment for Meta's Oculus headset - all tasks we have never taken on before!

What's next for InterviewsRUs

Upon completing this hackathon, we would really like to take this project a step further by further refining our VR environment for the user, mainly the graphics of the environment itself. Additionally, we would love to see this program utilize a speech to text feature: where a user can speak into a microphone, and the AI can interperate what the person has said in order to provide it's own response. Ultimately, we would love to see this application look and feel as close to the real world as our current technology allows.

Built With

Share this project:

Updates