Clearly - Automated Onboarding System

Inspiration

We wanted to solve a problem where we wanted to reduce the amount of time to do trivial and repetitive tasks for companies that have a social impact to the world.

What it does

Our project does the following things:

A dashboard consisting of all the candidates and their stage process within the onboarding system
Sends an email to our candidate for their referral with a tokenized link
Calls the referral using an AI voice agent
Summarizes the call transcript using Gemini AI and decides whether the candidate is suitable enough
Sends them a final email deciding the decision process.

How we built it

We used a variety of API's and ElevenLabs for the voice to create our project. We used Resend API to send emails for verification, Vapi to handle the agent calling, and Gemini to handle the call transcript. Furthermore, we created an elegant dashboard outlining the stage process. We store everything on MongoDB with the referrals associated to the candidate.

Challenges we ran into

The challenges we ran into consisted of dealing with the various API's and getting them to work together.

Accomplishments that we're proud of

A fully working agentic calling system that calls the referral and asks them referral questions. A dashboard that can show what stage it's at and does not need manual intervention.

What we learned

To work with various API's and work with new systems like Vapi.

What's next for Clearly - Automated Onboarding System

For Clearly, next we plan to implement an automated background search that verifies the applicant's information and flags any criminal charges. We also plan to implement a login system for security and to allow employers to work with our application.