Screen Smart

The app is an intelligent pre-screening solution that screens candidates from LinkedIn & other social/public platforms to find the fittest candidates. The app addresses the long hours recruiters put on unwanted/suspicious profiles for the given positions.

Our Inspiration:

Our talent recruitment team is overwhelmed with the higher number of openings, and they have an uphill task of shortlisting suitable candidates for the interview. Recruitment is a manual, time-consuming task where the recruitment & the interviewing team have to spend long hours shortlisting a candidate. On top of that, post-COVID pandemic, our recruitment team found 25% of profiles suspicious or fraud on different sources of data on the internet for each position, hence spending an average of 20+ hours a week on profiles which should ideally be rejected in the first place.

 

Our Solution:

This problem pushed us to think of any possible way to find the best profiles for the interviews in an automated way to find more appropriate talent in a lesser time. At the same time, a way to identify the suspicious profiles early in the screening process.
We developed this application on Appian that empowers the recruitment team to automatically enrich the candidate profile from multiple sources like LinkedIn, Twitter, and GitHub. The data collected is then analyzed intelligently to check if the profile matches the weighted match criteria defined in the application. This way, we get the results that are more suited to our needs.
The application brings the concept of Match Score, which is an easy way for the recruitment team to identify the right profiles for a job. The match score is an intelligently calculated score by the app based on how well the candidate profile matches the established criteria and how much it is cross-validated from multiple other data points. The Match score keeps on changing as more and more data is associated with the profile.
Finally, the applications also allow the recruitment team to perform virtual pre-screening assessment interviews. The application offers easy-to-use configurations for the recruitment team to specify the questions they would like to ask. Usually, these questions are the ones that can elicit more details about the candidate's profile, skill set. This video content is used to verify the skills identified from the unified data of LinkedIn, GitHub, and Twitter. If there is a mismatch of the user photo (for example, the LinkedIn photo does not match the Video assessment photo, etc.). The same is flagged as a suspicious profile.

Technology Stack:

1) Appian - core platform for building the entire application and connecting with multiple other technologies
2) Phantom Buster - for ease of integration with LinkedIn, Twitter, and GitHub.
3) AWS Lambda - to run core extraction/ ingestion logic runs on the scraped data from LinkedIn.
4) AWS S3 - for storing different profile pics, video recordings 
5) AWS Transcribe - converts the video speech into text.
6) AWS Rekoginition - to compare the profile images.
7) AWS Comprehend - for sentiment analysis, entity, and key phrases identification, etc. 
8) AddPipe - provides a ready-to-use platform for live video recording

Anonymous
Parents
  • I’ve been exploring for a little bit for any high-quality articles or blog posts in this kind of house. While exploring Yahoo I ultimately stumbled upon this website. Reading this info So I am glad to show that I have an incredibly good uncanny feeling I found out just what I needed. I such a lot indubitably will make Estate Macey White Blazer certain to do not to disregard this website and give it a look on a constant.

Comment
  • I’ve been exploring for a little bit for any high-quality articles or blog posts in this kind of house. While exploring Yahoo I ultimately stumbled upon this website. Reading this info So I am glad to show that I have an incredibly good uncanny feeling I found out just what I needed. I such a lot indubitably will make Estate Macey White Blazer certain to do not to disregard this website and give it a look on a constant.

Children
No Data