Netflix Data Scientist Interview Questions
& Process
Real candidates share what happened, how many rounds they had,
and how the experience turned out.
Based on 32 interview experiences · FREE TO READ
Candidate interview experiences
First-hand accounts from people who interviewed at Netflix.
Netflix
Data Scientist
Mid Difficult Positive experience No offer 1 round
Interview process
Technical screen
Interview formats
Technical
Start with regression related question, giving a detailed linear regression model. ( like explain the coefficient, R^2, type 1 & type 2 error) A/B testing, like if there is multi-class testing, what's the common pitfall? What's FDR, how to calculate it.
Confirmed questions1 question
- - how to explain R^2 in a regression model.
Netflix
Data Scientist
Mid Easy Neutral experience No offer 1 round About 3 days
Interview process
Onsite
Interview formats
Behavioral Technical
I really thought the interview went well. Or so I believed. The director seemed impressed, and so did the manager and recruiter. Netflix is expanding into other markets and they liked the experiences I have. The position is with the analytics department, which is more horizontal than vertical. I would have loved to work in a department like that. I have a lot of experiences with rec eng/sys and I seriously thought that I had a chance. Thinking about it now, I feel that they are prying for what I do in my current company, like methods, algorithms etc. Ideas are dime and a dozen. Good thing is I have more in the bag.
Confirmed questions2 questions
- Explain in more detail the rec engine you built for your company.
- Why are you using the XYZ method, and can you explain how you would improve content optimization?
Netflix
Data Scientist
Senior Average Negative experience No offer 1 round About 2 months
Interview process
Onsite
Interview formats
Behavioral Technical Coding
The scheduling was quite delayed with lots of communication gaps. I didn’t even know my onsite was scheduled until the hiring manager offered a prep call. The only positive part of the experience was the HM, who was proactive and responsive. During the onsite, the interviewers didn’t introduce themselves beyond just saying their name. Some of their statements about modeling sounded questionable. There was a lot of deep domain knowledge needed for most questions, even though it wasn’t listed as a requirement. I performed poorly with SQL, which was my own fault, and I also hadn’t prepared enough for white-board coding, which led to a no-go.
Confirmed questions2 questions
- What do you think is the best way to present ML results to stakeholders – the interviewer for some reason seemed to think RMSE was the best!
- When you split a population for an A/B test, what are some reasons you might see a significant difference between the control and the variant groups?
Netflix Data Scientist Interview Questions
Quoted word for word from Netflix interview reports.
“Write down the equations for building a classifier using Logistic Regression”
Read reports →“Q: Implement a Python method that returns the confidence intervals around a mean?”
Read reports →“- how to explain R^2 in a regression model.”
Read reports →“If you only had one day of data but with a large sample size and a significant test result, would you stop the experiment?”
Read reports →“If you are given a month of Netflix login data, including account_id, device_id, and payment-related metadata, how would you go about detecting fraud like identity theft or payment fraud?”
Read report →“How do you figure out if one algorithm is better than another one?”
Read report →“When you split a population for an A/B test, what are some reasons you might see a significant difference between the control and the variant groups?”
Read report →“How would you design and validate a metric to compare two users' ranked lists of movie/TV show preferences?”
Read report →“What do you think is the best way to present ML results to stakeholders – the interviewer for some reason seemed to think RMSE was the best!”
Read report →Formats, difficulty and experience
Across all 32 Netflix interview reports.