I applied through university. The process took 1 day. I interviewed at IBM (Ottawa, ON) in Oct 2009
Interview
I got the interview through the CO-OP program at my school. Interviewer was very nice, got along very well. I was able to answer most of the questions, none of them very technical. The most technical question was "What are the 3 main terms in a SQL Select statement". Questions were mostly about my past experiences, previous work terms, my classes etc. some hypothetical questions. Overall, pretty easy, just be yourself and be able to elaborate on your past experiences and be able to explain how you solved problems.
Interview questions [2]
Question 1
What are the three statements in an SQL Select statement?
Got a link to a platform where you were asked to answer a series of questions. The answers had to be recorded, every answer was supposed to be answered in less than 3 minutes. Beforehand, the advice was given to prepare using the STAR framework.
Important: once started, you can only interrupt the process once, since they don't want you to have unlimited preparation time for the answers. If it gets interrupted (e.g. due to network connectivity or by closing the tab) a second time, the process is stopped and your interview is basically aborted.
Interview questions [2]
Question 1
Tell us about a time you disagreed with a coworker and how you resolved the disagreement.
First step involved an easy coding assessment, then 2 more rounds of online interviews. Overall process was quite slow until confirmation but talent acquisition team was helpful and responsive in guiding the process.
Interview questions [1]
Question 1
Say an AI system works like this, but there's an accuracy issue. What measures would you take to improve system accuracy.
60 minute technical interview was pretty chill, just asked about projects and basic machine learning concepts. Interviewer was very nice, didn't even take the full 60 minutes. There was also an OA at the beginning, was like an easy to medium leetcode on Hackerrank.
Interview questions [1]
Question 1
What are ways of evaluating performance in a machine learning model?
