Oliver Wyman Data Scientist Interview Questions
& Process
Real candidates share what happened, how many rounds they had,
and how the experience turned out.
Based on 10 interview experiences · FREE TO READ
Candidate interview experiences
First-hand accounts from people who interviewed at Oliver Wyman.
Oliver Wyman
Data Scientist
Mid Easy Positive experience No offer 3 rounds
Interview process
Technical screen Onsite Recruiter call
Interview formats
Technical Case Behavioral
It was three rounds. Round 1 was about basic ML resume stuff and case studies related to ML (like how to get features, pre processing, training, testing). Round 2 was a pure case study (I'd suggest looking at IIM case books for each type of case and structure) and sometimes you might get guesstimates. Round 3 was mostly HR oriented.
Confirmed questions7 questions
- What assumptions does logistic regression make?
- Can you explain the assumptions behind Random Forest?
- What are the assumptions for random trees?
Oliver Wyman
Data Scientist
Lead Average Neutral experience No offer 1 round
Interview process
Recruiter call Technical screen
Interview formats
Technical Behavioral
I interviewed with an Associate Director who was recently hired, a level above the role I applied for. They seemed to have some knowledge of AI/ML from online courses, asking basic questions initially like why accuracy isn't always the best metric and what a p-value is. By the end, it was clear they lacked practical experience in deploying models. Their follow-up comments suggested I might have been a better fit for the Associate Director role, perhaps because I wasn't as hands-on at the time, which is why I didn't get a callback.
Confirmed questions3 questions
- Can you explain why accuracy isn't always the best evaluation metric, and in which scenarios recall would be preferred?
- What is a p-value?
- What are the differences between LightGBM, Gradient Boosting Machines, and XGBoost?
Oliver Wyman
Data Scientist
Senior Average Negative experience No offer 5 rounds
Interview process
Recruiter call Technical screen Take home Onsite Panel
Interview formats
Behavioral Technical Coding Case Presentation
The interview process was unnecessarily long, spanning 5 rounds over 2-3 months. Round 1 was with HR for a behavioral fit. Round 2 involved a technical interview with the Lead Data Scientist, covering common data science questions and a project discussion. Round 3 was a take-home assignment focused on credit risk modeling, followed by a discussion. Round 4 consisted of a credit risk case study discussion. Round 5 was a partner round involving a project and a puzzle. I made it to the final round and felt I performed well, even solving the puzzle, but was unfortunately rejected for reasons I don't understand, related to my narration and puzzle-solving approach. It felt like a waste of time.
Confirmed questions8 questions
- Tell me about your motivation for seeking a new job.
- Can you write SQL queries involving CTEs and window functions at a medium difficulty level?
- What are some fundamental data science concepts?
Oliver Wyman Data Scientist Interview Questions
Quoted word for word from Oliver Wyman interview reports.
“What are the assumptions for random trees?”
Read reports →“What is a p-value?”
Read reports →“What assumptions does logistic regression make?”
Read reports →“If machines fail 1% of the time, how big should a cluster be to ensure at least one machine fails daily?”
Read reports →“What are the differences between LightGBM, Gradient Boosting Machines, and XGBoost?”
Read report →“Can you explain the assumptions behind Random Forest?”
Read report →“Can you explain Random Boost and its difference from linear and lasso regression?”
Read report →“Can you explain why accuracy isn't always the best evaluation metric, and in which scenarios recall would be preferred?”
Read report →“How do you validate the performance of an ML model?”
Read report →Formats, difficulty and experience
Across all 10 Oliver Wyman interview reports.