HireIQ

HireIQ automates early-stage recruiting with lifelike interviews, eye-tracking authenticity checks, and instant AI evaluations

Comment

Inspiration

We wanted to create a service that not only serves people but offers a win -win solution to both sides using it. One of the biggest problems today is the job market. Hiring managers and recruiters have to go through hundreds of thousands of application to hire people, and on the other hand, people are applying to several hundred jobs without hope of even an interview for weeks. This is when we thought we could bridge this gap using our knowledge and skills.

What it does

HireIQ revolutionalizes the application process for both, the applicant and the employer. When an employer posts a job on the app, potential candidates can then see this and if they think they're a good fit for the job, they can apply solely through their resume, which will match their resume to the job description based on similarity of keywords. On a match of more than 85%, the applicants instantly receive an email with an online assessment to get to the next part of their application. On successful completion of the online assessment, the applicant receives a behavioural interview conducted by an AI avatar that is prompted by the employer based on desired qualifications, and powered by eye tracking system to prevent cheating. If a candidate has passed all these tests, they then can meet with the hiring manager for a final round of interview to decide whether they are hired or not.

How we built it

We built our project as a full-stack AI job application platform designed to make hiring faster, fairer, and more engaging through natural, human-like conversations. We used FastAPI to make routing smoother, Gemini to handle resume matching, generate interview questions based on resume, give scores based on transcripts and summarize the interview. We used Snowflake for database management. We used Judge0 which is an open source software to conduct the online coding tests. We used eye tracking to give a integrity score to the recruiter and also to prevent cheating during the online coding tests. SendGrid to send emails, if the candidate matches the resume with more than 85% accuracy. Additionally, we used d-id to develop an AI agent which we fed technical data and resume's to, so that it conducts realistic interviews. For front-end we used next.js as we were both familiar with it and it has tons of libraries for a modern web application.

Challenges we ran into

Me and my teammate we ran into a problem where the code would not detect the gemini API key which took us a long time to debug and finally arrive at a solution. Initially we were planning on using elevenlabs and d-id to build an even more realistic AI recruiter, but soon we realised that d-id was not able to access the audio files generated by elevenlabs, So we had to improvise and we came up with an idea of building an AI agent out of d-id.

Accomplishments that we're proud of

We are really proud of the online assessment platform which has an in-built eye tracker system preventing cheating and promoting fairness. Moreover, we are also proud about the AI agent that we fed technical data to and made it a very realistic interviewer with high-knowledge in the field of tech.

What we learned

One of the most interesting things we learnt was the judge0 we tried really hard initially trying to make it by ourselves, and then we came across judge0 which helped us a lot. We learnt how to efficiently prompt AI agents and feed them data to conduct interviews with minimal scope of error

What's next for HireIQ

We plan to take HireIQ to the next level by making interviews even more dynamic, insightful, and scalable. Me and my teammate would like to take this forward and make this idea come to life as we believe it has immense potential in it.

Built With

Share this project:

Updates