LinkedIn Data Scientist Interview Questions
& Process
Real candidates share what happened, how many rounds they had,
and how the experience turned out.
Based on 87 interview experiences · FREE TO READ
Candidate interview experiences
First-hand accounts from people who interviewed at LinkedIn.
LinkedIn
Data Science Engineer
Mid Difficult Negative experience No offer 3 rounds
Interview process
Recruiter call Technical screen Onsite
Interview formats
Technical Coding Behavioral
The interview process was really bad and rude, from HR to the interviewers. The HR part was particularly weird; after I gave my availability, it took about a week to get a response. When I followed up, they said there were issues with the recruiting coordinator and asked for my availability again. After the onsite interview, there was no update on the results via call or email until I reached out. It's surprising this happened at LinkedIn. Also, why was a mid-level position interviewed by a staff-level person instead of senior folks? During the coding questions, after I quickly solved the first one, the interviewer seemed suspicious about the second, saying I 'may have met the first question before' when I spent more time on it than expected. I was quite angry about that comment. Overall, the entire experience was quite awful and nonsensical.
Confirmed questions4 questions
- What are your thoughts on Leetcode array, string, or sliding window problems?
- Regarding statistics, how would you determine the probability of rolling dice?
- Can you tell me about the most challenging project you've worked on?
LinkedIn
Data Scientist
Mid Easy Positive experience No offer 1 round
Interview process
Recruiter call Technical screen
Interview formats
Technical Coding Technical
After the initial intro call, I was invited for a coding and ML interview. It had two parts: first, a coding question for about 25 minutes where I had to implement an algorithm to sample from a discrete distribution using basic random generators. I solved it by creating intervals on the real line based on probabilities and using binary search. Second, they asked foundational ML questions like explaining logistic regression, its loss function, and optimization, plus questions on evaluation metrics and methods for ML algorithms.
Confirmed questions1 question
- How would you implement an algorithm to simulate rolling an M-sided dice given the probabilities of each side?
LinkedIn
Data Science Manager
Manager Average Negative experience No offer 4 rounds
Interview process
Recruiter call Phone screen Onsite
Interview formats
Technical Case Behavioral
The interview process had so many phone interviews. They seemed to be deliberately delaying the process, as if they wanted to interview everyone in the Bay Area before deciding. They also took a long time to respond between interviews, which made me feel they aren't meritocratic but full of biases and egos. The recruiter for the DS manager role was very rude and tried to establish herself by rudely responding to candidate emails. The hiring manager chat was great, and I was sure she was impressed. However, sometimes you're not chosen because you're too good and might be a threat later. Glad they rejected me, though; they're looking for a soldier, not a leader, despite it being a leadership role. The data science team is very homogenous, only accepting one type of profile, and doesn't seem to welcome diversity in skills and appearance.
Confirmed questions4 questions
- Questions about SQL
- A business case scenario
- Questions about past experiences and behavior
LinkedIn Data Scientist Interview Questions
Quoted word for word from LinkedIn interview reports.
“Create a function that generates a uniform random number between 1 and 7, using a generator that produces numbers from 1 to 5 uniformly.”
Read reports →“Given an array, find an index to start at such that the sum is never negative.”
Read reports →“How would you segment a long string into valid words using a dictionary, returning false if it's impossible?”
Read reports →“Can you formulate the L1-norm minimization of a linear system residual as a linear program?”
Read reports →“Is L1 Lasso or L2 Lasso? And which one is Ridge regression?”
Read report →“How would you find the k most frequent numbers from a stream of numbers in real-time?”
Read report →“How would you go about generating a sorted vector from two given sorted vectors?”
Read report →“Given a table of page view data with Member ID and timestamp columns (representing a 24-hour period), and knowing that members and guests/visitors are distinguishable by ID, how would you identify crawler-related activities (excluding my members) and what would the query look like?”
Read report →“How would you implement an algorithm to simulate rolling an M-sided dice given the probabilities of each side?”
Read report →Formats, difficulty and experience
Across all 87 LinkedIn interview reports.