Uber Data Scientist Interview Questions
& Process
Real candidates share what happened, how many rounds they had,
and how the experience turned out.
Based on 232 interview experiences · FREE TO READ
Candidate interview experiences
First-hand accounts from people who interviewed at Uber.
Uber
Data Scientist
Intern Difficult Positive experience No offer 1 round
Interview process
Onsite
Interview formats
Behavioral Technical
The interviewers were very professional. The questions were detailed and centered on understanding processes and clearly explaining concepts to people without prior knowledge of the topic. They were open to questions and happy to provide guidance and advice.
Confirmed questions1 question
- Bias-variance tradeoff, p-value, how to do train test split on time series.
Uber
Data Scientist
Mid Average Positive experience No offer 5 rounds About a month
Interview process
Technical screen Technical screen Technical screen Panel Panel
Interview formats
Technical Case Other
5 rounds of interview: 1- HM round with 15 min CV discussion, 2 to 3 sql questions on grouping and joins and then a short 15 min case study about Fraud ride cancellation. 2. 1 hour round for SQL and pandas. SQL is joins, windows function ranking lead lag . 3. HM round for other team. CV discussion and A/B TESTING Case study based on previous work experience. 4. PM round mainly focused on Case study- RCA type case study to find the drop in a metric for an experiment launched. 5. Director round- Again short intro and directly went into case study to measure product success for a feature/fraud prevention product launch. Last round was fully focused on structure and strategic thinking. Wasn't able to perform well. Rejected
Confirmed questions5 questions
- Round 1 sql + choosing metrics for fraud detection
- Round 2 advanced sql round
- Round 3 case study plus work experience discussion
Uber
Data Scientist
Senior Average Negative experience No offer 2 rounds About 7 days
Interview process
Technical screen Technical screen
Interview formats
Technical Case
The process had two screening rounds (one focused on SQL and another on Analytics case study/A-B testing knowledge). The first round went well, all SQL questions were answered promptly. The analytics case study was about a problem statement and how an A-B test could be designed to solve that, which I was able to solve and answer. Two days later HR came back with negative feedback. I asked if they could share detailed feedback, and HR replied that they only received a simple no from the interview panel (for both rounds) and had to follow up with the panel for detailed feedback. I followed up multiple times with HR to understand the gaps (was very surprised by the feedback on the SQL round) but got no response. I still have no idea what kind of expectations the interview panel had or if they were conducting the interview rounds just to reject candidates. They need to respect the time the applicant is investing in preparing for these rounds. The least they can do is provide actionable feedback so the candidate can introspect.
Confirmed questions1 question
- Simple questions on SQL lag, lead, joins, dense_rank, and CTEs
Uber Data Scientist Interview Questions
Quoted word for word from Uber interview reports.
“Given a random Bernoulli trial generator, implement a function that outputs a value sampled from a normal distribution.”
Read reports →“Write a production-level code to get all combinations of numbers in a list that add up to 8.”
Read reports →“Implement a text wrapper, for example split a long sentence by a character limit but only break at spaces.”
Read reports →“(Stochastic process problem) What are the expected time units until you first get H,T,T when flipping a fair coin?”
Read reports →“Write code that simulates the roll of a die, assuming you only have a continuous uniform random number generator.”
Read report →“The data shows drivers earn more per hour when they go to neighborhoods near an airport instead of waiting at the airport for a new ride. So why are drivers still more likely to stay at the airport?”
Read report →“If recall is equal some number and precision is equal another number, what should be the values in the precision matrix?”
Read report →“If we add one more rider to the current SF market, how would that impact the existing riders and drivers?”
Read report →“What kind of trends in the data would show you that the market is healthy? What is price telling you?”
Read report →Formats, difficulty and experience
Across all 232 Uber interview reports.