My Cousin is an HR exec and she always complained about poor aptitude along with other lacking traits of potential candidates a few steps further in the interview process.
This lead me to build GenomeHR that scans images (lets say from LinkedIn) and uses GENOME LINK api to get insights (PHENOTYPES) such as
- Childhood intelligence,
- Mathematical ability
- Openness . .
etc to scout for better candidates.
What it does
AI Module classifies candidates by ethnicity ------> *GENOME LINK uses this to predict the "Smartness" & eligibility of candidates *
How I built it
Uses a Tensorflow image model to predict the persons ethnicity. Uses GENOME LINK api to get traits Python web app to showcase the product.
Challenges I ran into
Training data and training the module was challenging to achieve accuracy
Accomplishments that I'm proud of
The AI / ML model's accuracy
What I learned
Genome api functionalities. How to build image recognition models.
What's next for GenomeHR
This can be used as a tool / browser extension that activates when HR personnel are scouting on Linkedin or any other website of similar kind
Can be used to train the model specific to organizations.
Can be used to identify employees who potentially could be recommended -mathematical programs -reading programs -anger management programs -mental health programs.
Log in or sign up for Devpost to join the conversation.