The following incident happened at IIT Madras. I was interviewing students for software engineering positions.
Students would enter the interview room, receive a question, and try to solve the question in 45 minutes.
While they did this, interviewers and HR would communicate on Slack. This helped us resolve tough questions from the interviewees as a group 😛
As the interviews progressed, we updated a spreadsheet. Against every applicant's entry, we had a status and reason.
Interviewee performance and status spreadsheet.
After 2 hours, we got together to debrief. Here is a short gist of our conversation:
I1: This applicant looked promising, why did you reject them?
I2: He was good at problem-solving, but not at coding.
I1: My candidates were good at coding but gave brute-force approaches to the problems.
I2: If we had to rank these candidates, how would we do so?
I1: Why would we rank them? This is a filter, pass or fail. If you aren't good at everything, we reject you.
I2: What about second chances? If we had to give an average score, how would we do so?
I1: Let's give them a score out of five.
I2: That's a good idea, but a general score is vague. Let's give them scores on coding, problem-solving, and communication.
I1: Great. So we score them on a total of 15, and rank order them. -- And we give second chances to people with scores greater than 9.
I2: Awesome!
That's when we understood the importance of a scorecard. It makes the evaluation objective, keeps the conversation directed, and reduces interviewer bias.
Big tech companies nearly always use a scorecard for interviews. Here's a sample for system design interviews.
System Design Interview Scorecard.
In the next blog, we will explain how this scorecard or "competency scale" works, and what it means for job aspirants.
See you in part 2!