
SimilarWeb Data Scientist Interview Questions
& Process
Real candidates share what happened, how many rounds they had,
and how the experience turned out.
Based on 15 interview experiences · FREE TO READ
Candidate interview experiences
First-hand accounts from people who interviewed at SimilarWeb.
Data Scientist
The interview process started with a recruiter call, followed by a team interview that included a coding assignment, and then a case study. Honestly, it was quite a bad experience. The recruiter seemed disorganized and often took two weeks to respond, requiring multiple follow-ups. In each call, they weren't clear about my status or what I'd already completed. After submitting the case study, the feedback was simply that it "wasn't good enough." Despite multiple requests for more detailed feedback, I was ignored. The level of unprofessionalism was astonishing.
- Perform a clustering task to group similar websites.
Data Scientist
Had a good overall experience. The whole thing took about 2.5 weeks, which was fine since I wasn't in a hurry. Started with a quick 20-min HR chat, then a 20-min intro call with a team member where they asked a technical question. After that, there was a 1.5-hour technical interview with two team members. Finally, I had back-to-back onsite interviews with different people from the department and HR. I really appreciated how the recruiter stayed in touch throughout, checking in on how I was doing and giving updates on what was next. The take-home assignment was also pretty good, not like a huge extra project, and felt respectful of my time.
- Given that the probability of a single match being defective is p and the probability of a matchbox being defective is q, if we attempted to light a match and it failed to ignite, what is the likelihood that the subsequent match we try will ignite?
Data Scientist
The process started with the usual "tell me about yourself" question. Then, the interviewer focused solely on probability questions and asked me to write algorithms based on probability case studies, like the example provided.
- Could you start by telling me about yourself?
- Imagine a maze with three paths, only one of which is correct. How would you design an algorithm to ensure an equal probability of selecting each path on every attempt?
SimilarWeb Data Scientist Interview Questions
Quoted word for word from SimilarWeb interview reports.
“Given that the probability of a single match being defective is p and the probability of a matchbox being defective is q, if we attempted to light a match and it failed to ignite, what is the likelihood that the subsequent match we try will ignite?”
Read reports →“Could you write a function that takes an array and returns an element from it, based on a provided function that generates a number between 0 and 1 with uniform distribution?”
Read reports →“Can you explain and implement the KNN algorithm?”
Read reports →“Can you tell me about polynomial linear regression and overfitting?”
Read reports →“Imagine a maze with three paths, only one of which is correct. How would you design an algorithm to ensure an equal probability of selecting each path on every attempt?”
Read report →“What are your understandings of statistical concepts?”
Read report →“How did you verify the validity of your project implementations?”
Read report →“How would you approach theoretical problems in real-life scenarios?”
Read report →“Could you start by telling me about yourself?”
Read report →Formats, difficulty and experience
Across all 15 SimilarWeb interview reports.