State Farm Data Scientist Interview Questions
& Process
Real candidates share what happened, how many rounds they had,
and how the experience turned out.
Based on 68 interview experiences · FREE TO READ
Candidate interview experiences
First-hand accounts from people who interviewed at State Farm.
State Farm
Data Scientist
Mid Average Negative experience No offer 5 rounds
Interview process
Take home Recruiter call Technical screen Technical screen Technical screen
Interview formats
Technical Presentation
They make you jump through hoops with at least 5 steps. You don't even talk to a real person until the 4th step. They make you waste hours with obscure or generic questions, unsure about what or why they're asking, and you have to film a response. If you pass that, you can actually talk to a recruiter who tries too hard to sell how great the company is and sets up multiple more rounds of 'technical' interviews. These were painful and a waste of time. They are overly specific about what they already do and their teams and practices but won't share what or why. The interviewers only seem focused on showing they are smarter than you. After eventually getting rejected, I got feedback from the same interview of 'didn't dive into details only explained overlying concepts' and 'only focused on details and did not explain the how or why,' completely useless and neither of which were prompted when I asked if they had any questions about my answers. Communication is a 2-way process and they clearly fail.
Confirmed questions2 questions
- Why are you interested in working here?
- Can you explain gradient boosted logistic regression, specifically detailing the algorithm and any process deviations for implementation?
State Farm
Data Scientist
Entry Average Negative experience Accept offer 4 rounds
Interview process
Recruiter call Take home Onsite Panel Background check Offer
Interview formats
Technical Coding Behavioral
The entire hiring process spanned approximately two months. Initially, I encountered a Hirevue assessment with straightforward questions on topics like the bias-variance trade-off. About a month later, I had a call with HR to confirm details and was given a dataset to analyze. This dataset was challenging due to significant class imbalance. After submitting my work, the final round was scheduled for the following week. This involved a two-hour session with three different management groups, focusing primarily on behavioral questions. Notably, I was asked twice to detail my methodology for dataset analysis, requiring a comprehensive explanation of my entire process. While the technical questions weren't overly complex, a solid understanding and the ability to elaborate on each step, justifying model choices and their advantages, were crucial. The process concluded with a meeting with the department director, who inquired about my background and asked further behavioral questions in a patient manner.
Confirmed questions5 questions
- Can you explain the trade-off between bias and variance?
- Could you walk me through your typical workflow when tackling a dataset?
- Explain how you would approach a dataset and detail the entire process.
State Farm
Data Scientist
Mid Difficult Positive experience No offer 3 rounds
Interview process
Take home Technical screen Onsite
Interview formats
Case Technical Presentation
So first, I had a case study with some data, it was a classification problem where I had to use logistic regression and a non-generalized linear model algorithm. I also had to create an executive summary for the models. Then, I had a technical interview with 2 recruiters where we went over the case study, discussing methodologies, assumptions, and reasonings. Finally, there was an on-site interview with different teams and the hiring manager.
Confirmed questions5 questions
- Could you tell me what the link function is for logistic regression?
- What regularization method does sklearn use by default for logistic regression?
- Can you elaborate on the methodologies, assumptions, and reasoning behind your case study approach?
State Farm Data Scientist Interview Questions
Quoted word for word from State Farm interview reports.
“Given a 95% confidence interval, the sample mean claim amount is $175 in Georgia and $200 in Illinois. If the margin of error is $15, which location has a higher cost and why?”
Read reports →“If we have the linear regression equation y=10+5X and an R^2 of 0.65, what does this R^2 value indicate?”
Read reports →“For the linear regression equation y=10+5X with an R^2 of 0.65, what does this R^2 value mean?”
Read reports →“Given the linear regression equation y=10+5X and an R^2 of 0.65, will the coefficient value increase, decrease, or stay the same if the data's units change from days to years?”
Read reports →“Considering the linear regression equation y=10+5X and R^2=0.65, if the data's unit of time changes from days to years, will the R^2 value increase, decrease, or remain the same?”
Read report →“If a logistic regression model's top feature from a random forest selection wasn't statistically significant, what are at least two possible reasons for this?”
Read report →“Consider a scenario where a tree-based model identifies 20 important features, but a Logistic Regression model doesn't find the top variable statistically significant. What could explain this discrepancy?”
Read report →“Did the predictions represent the class probabilities for belonging to the positive class (labeled '1')?”
Read report →“Was a prediction generated for each of the rows in the test dataset (10K rows)?”
Read report →Formats, difficulty and experience
Across all 68 State Farm interview reports.