You are on page 1of 2

Introduction:

The pandemic has been ongoing for last couple of years & has required most of the organizations to
adapt to the changing landscape and come up with innovative ways to identify talent and recruit them.

As the number of employees to be hired increases, the total time spent on hiring impacts overall
business.

The Challenge:

There is a need to arrive at identifying the right talent for a role in a more automated and reliable
manner.

Introducing Machine Learning to the area of candidate selection would help the industry to come up
with mechanisms to reduce the amount of time spent by organizations to find the right fit candidate and
also help the candidates find their dream role.

Build a Talent Matching engine which based on the role, which identifies the probability of the candidate
being selected. Typically candidate might have 3 rounds of Interview and can drop at any stage &
feedback needs to be captured at different stages.

The system built should be extensible to adapt to changing requirements and be scalable and secure.

Be as innovative as possible to arrive at the solution. Feel free to address the problems in your own way
and use following points merely as guidelines.

While the resume matching algorithm is at the core, additional requirements to make the solution
viable are as follow:

1. A login page for candidates and a login page for hiring manager.
2. A form where the candidate can input their information points based on the schema defined
below.
3. A form where the roles can be input by the hiring manager based on the schema defined below
4. A form where the roles status can be input by the hiring  based on the schema defined below
5. A page where a hiring manager can identify the candidates which best fit in the provided role.
6. Validation: Reject Duplicate Candidate insertion by one of the candidate information criteria
(Mobile/email)
7. Validation: The candidate can only be interviewed at one role at any point of time. Once
selected the candidate is dismissed for selection from roles
8. Optional: A Scraper which scrapes through provided resume (instead of forms assume word
doc/pdf) and inserts in the table
9. Optional: Think in terms of providing utilities like bulk upload/ bulk select/ etc for the
candidates/hiring team
10. Optional: Capture whether the selected candidate has joined. If dropped after making an offer –
Flag the candidate Drop with Reason and Also the timeframe before the joining date the
candidate dropped. Over a period reports like – Are we seeing rejection from a particular
Interview, Skill or LOB, Tenure.
Assumptions:

The below database schema can be used as a reference point for the information point to be captured
for building the recommendation engine

1. Candidate Information :
Candidate Id, Candidate name, Description , Years of experience , Mobile #1,Mobile #2, Email ,
Date Of Application , Preferred Location, Additional skills, Current Role, Current Organization
2. Candidate Skillset:
Candidate Id, Skill Set ,Skill rating, Skill Experience(In Years), LOB
3. Roles Information:
Role Id, Role Name, Description, Skill Set, Years of Experience, Group, Hiring Role
(Analyst/Assoc/VP) , Location, Hiring Manager
4. Role Status:
Role Id, Status (Rejected, On Hold, Approved, Ongoing, Dropped), Candidate Id , Rejection
Reason 1, Rejection Reason 2, Rejection Reason 3 , Drop Reason 1 , Drop Reason 2 , Drop Reason
3, Candidate Strength 1, Candidate Strength 2, Candidate Strength 3, Round of Rejection;
Interviewer ID and Name

You might also like