Spotify Data Scientist candidates report a recruiter screen, a SQL/Python/statistics technical round, a hiring manager conversation, and in some cases a case study or presentation, with a strong A/B testing and applied-judgment focus.
$165K
Avg. Base Comp
$200K
Avg. Total Comp
3-5 rounds
Typical Rounds
Not reported
Process Length
Candidates who describe Spotify's Data Scientist process report a technical core of SQL, Python and statistics, followed by a hiring manager conversation that tests applied judgment. Reported SQL topics include aggregations, joins, date filtering, ranking and window functions such as LAG. Python questions range from FizzBuzz to Pandas filtering and conditional updates. One candidate also saw a Three Sum-style coding problem.
Experimentation comes up repeatedly. One candidate had to interpret an A/B test result and turn it into a business recommendation, not just a statistical conclusion. Another described a cold-start experimentation case about launching a new product line. A third said A/B testing concepts came up across rounds.
The hiring manager round appears to be where candidates are pushed hardest. One candidate was questioned on why they chose a methodology, what limits the data had, and whether findings could generalize. Another was asked a business question about what Spotify should pay a platform for Premium ads, using CLTV and CAC. Two candidates also mention a case study or presentation stage, with one prompt about what makes a playlist successful.
Reports differ on structure and outcome. Candidates describe three to five stages, and the reported outcomes were rejections plus one declined offer. Prepare to code live while explaining your reasoning, and bring project stories that show cross-functional impact and honest limitations.
Synthesized from 5 candidate reports by our editorial team.
Had an interview recently?
Share your experience. Unlock the full guide.
Real interview reports from people who went through the Spotify process.
Share your own interview experience to unlock all reports, or subscribe for full access.
Sourced from candidate reports and verified by our team.
Topics based on recent interview experiences.
Featured question at Spotify
Given an integer N, write a function that returns all of the prime numbers up to N
| Question | |
|---|---|
| Top 3 Users | |
| Decreasing Comments | |
| The Brackets Problem | |
| Valid Anagram | |
| Hurdles In Data Projects | |
| Banner Ad Strategy Success | |
| Third Unique Song | |
| Count Transactions | |
| Declining Usage After Launch | |
| A/B Testing a Checkout Button Change | |
| Data Preparation for Imbalanced Data | |
| Estimating D | |
| String Palindromes | |
| Three Indexes Adding Zero | |
| Confidence Interval Explanation | |
| Pathfinder in Maze | |
| Safe Deployments | |
| Check Matching Parentheses | |
| Minimum Directional Path | |
| Duplicate Product Names | |
| Ranking Metrics | |
| Generating Discover Weekly | |
| Podcast Space | |
| Singly Linked List | |
| Podcast Search | |
| Prime Music Integration | |
| Bootstrapping Samples | |
| Third Party Ad Pricing | |
| Empty Neighborhoods |
Synthesized from candidate reports. Individual experiences may vary.
Candidates report starting with a recruiter conversation. One said the recruiter explained the interview structure, which matched what followed. Another said the first call included motivation questions about the role and Spotify. Timing between contact and response may vary, and one candidate reported long silence before hearing back.
Candidates report live technical questions in SQL, Python and statistics. Reported SQL topics include joins, date filtering, aggregations, ranking, time spent and LAG window functions. Python ranged from FizzBuzz to Pandas filtering. One candidate also saw chart interpretation and inference, and another saw an A/B test interpretation. Syntax slips under time pressure were noted, and explaining your reasoning aloud appears to matter.
Candidates report a deeper conversation about past work. Topics included methodology choices, data limitations, whether findings generalize, and cross-team collaboration. One candidate got a CLTV and CAC ad-pricing business question, and another got a cold-start experimentation case. Prepare to defend your decisions and acknowledge limits plainly.
Two candidates mention a case study or presentation stage. One was asked what makes a playlist successful and was rejected after that round. The other described a final presentation focused on synthesizing findings for a broader audience. Not every candidate reached this stage, so it may depend on how earlier rounds go.
One candidate reports a final round with business stakeholders after the case study. Another said interviewers across the process cared about how they would collaborate with product, engineering and business teams. Treat this as a likely emphasis on collaboration rather than a fixed format, and be ready to explain results to non-technical partners.
