InterviewMe

Inspiration

Interviewing can be nerve-wrecking for everyone. Many resources capitalize on the difficulty of technical interviews by teaching fundamental concepts, but having good responses ready for the behavioral interview can be just as important. It can be inconvenient to schedule a mock interview with another person, or it might be uncomfortable narrating your personal experiences to someone else. That's where InterviewMe comes in.

What it does

Interviews are hard, but being good at them truly comes down to practice and confidence. InterviewMe gives users canonical behavioral interview questions, allowing them to record their response and receive meaningful analytics. Users can replay their response audio and evaluate it from the interviewer's perspective to improve their own speaking. They can also see whether they took too long, spoke too fast, used too many filler words, or did not hit all of their main points. These are all good practices that people tend to forget in the heat of an interview. But not to worry, because InterviewMe allows users to solidify their responses to these questions, making it less off-putting when they are asked the same types of questions in an actual job interview.

How we built it

The web app was built with the React-Django full stack. In the backend, Django establishes a REST API for calling interview questions and response analytics. The backend also calls to Google Cloud's Speech-To-Text API.

Challenges we ran into

None of us had experience with speech-to-text conversion, and Kai Chen was able to use Google Cloud's API to integrate functionality into our project.

We wanted to make the experience as beneficial as possible to the user, and it was difficult to uncover good metrics to analyze from a user's speech as text. We ended up evaluating time taken, words per minute, filler word count, and noun phrase extraction as results from which the user could extract value.

Accomplishments that we're proud of

None of us were full stack specialists, and we're glad that we were able to link everything together with the frontend and backend during the 36 hour window.

We also got good sleep during the hackathon as well, so it was a healthy experience!