Customer Facing Data Scientist applicants have rated the interview process at DataRobot with 3.1 out of 5 (where 5 is the highest level of difficulty) and assessed their interview experience as 40% positive. To compare, the company-average is 43.6% positive. This is according to Glassdoor user ratings.
Candidates applying for Customer Facing Data Scientist roles take an average of 55 days to get hired, when considering 15 user submitted interviews for this role. To compare, the hiring process at DataRobot overall takes an average of 32 days.
Common stages of the interview process at DataRobot as a Customer Facing Data Scientist according to 15 Glassdoor interviews include:
Phone interview: 35%
One on one interview: 18%
Presentation: 15%
Group panel interview: 12%
Skills test: 12%
Background check: 3%
Personality test: 3%
Other: 3%
Here are the most commonly searched roles for interview reports -
I applied online. The process took 2 months. I interviewed at DataRobot in Jul 2019
Interview
One word - 'Nonsense'
They keep you engaged in a 2 months-long interview process only to end up with a NO, citing a reason that they are picky in their candidate selection. Very well, if you are so picky you are supposed to figure the candidate skillset out in first two interviews itself.
The interview structure in itself does not make sense. They organize sequential interviews, one interview in 2 weeks, all on video chat as their interviewers are all at different locations. I cleared the first 4 rounds and then the 5th guy did not find me good. So, basically everything that I did in the first 4 rounds is useless? Is it some sort of video game where you go stagewise? You are keeping a candidate busy for 2 months and then, in the end, you say you could not cross this particular stage. No company does that. Every company takes a first telephonic interview and then if found fit, they organize one full day interview even if it is on a video call. And then make a decision based on combined feedback of all interviewers.
Basically, what I understand is that they are interviewing multiple people for one position in the 2 month period and then they pick the best one they find. Your time and effort do not matter to them.
Finding the best candidate is every company's right but wasting the time of candidates in the two-month-long process is not right.
Don't waste your time with them if you are actively seeking a job. It is good only for people who are passively looking for a job.
Interview questions [1]
Question 1
Questions on your projects and several machine learning algorithms
I applied through a recruiter. The process took 5 months. I interviewed at DataRobot (Dubai)
Interview
5 rounds some technical questions mostly questions about prior experience and asked to deliver a presentation about previously delivered project
No coding take home assignment, no leetcode, no real hard DS grilling questions more trying to gauge your ability to explain technical concepts both to technical and non technical audience members
Interview questions [1]
Question 1
Could you present to us something you've previously worked on
The process included several rounds of talking with first a recruiter, then members of the team I was joining. It culminated with a presentation interview to two current employees posing as customers
I applied through a recruiter. The process took 3 weeks. I interviewed at DataRobot (Washington, DC) in Nov 2021
Interview
Approximately 5 technical (virtual) interviews and 1 behavioral. The technical interviews were highly repetitive in asking about the same supervised machine learning algorithms, pros/cons, and troubleshooting steps. Felt like there was no continuity between interviewers. Technical interviewers were adequate (but not excellent) technical communicators, but they were looking for very specific answers. Once I hit some arbitrary threshold of answering a question correctly, they'd interrupt and fire off the next question.
Interview questions [1]
Question 1
Name common preprocessing (cleaning) steps for a supervised machine learning model.
How would you deal with severe class imbalance while developing an ensemble model?