intr.vu

Inspiration

The technical interview is overwhelming at times and many are unfamiliar with the landscape of the technical interview. Therefore, we made mock technical interviewer so it can help others improve with their technical interview experience.

What it does

It is a virtual interviewer that takes in speech of user to process it through a LLM and respond back to the user in a professional manner.

How we built it

We built it with Next.JS with React frontend and Express for the backend. We utilized OpenAI Whisper for the speech to text from the user, process it through GPT 4 and Eleven Labs text to speech for the response back to the user.

Challenges we ran into

Some challenges we ran into were the synchronization errors that we constantly hit when we had to import and use new libraries/tech.

Accomplishments that we're proud of

We are proud of getting the entire flow of speech to text and from text back to speech working as well as getting all the leetcode problems with tracking of user progress.

What we learned

We learn that even though some parts of the project may seem like it would work well together flawlessly, there may still be version issues or merge conflicts. Therefore, we should always keep our code with the future in mind and not mindlessly submit what we have to the codebase.

What's next for intr.vu

Scaling the website bigger, there can be more variety of questions, better user feedback after the interview and more incentives to make the experience more enjoyable for users.