United Airlines Data Scientist interviews, as candidates report them, mix project deep-dives, ML and statistics fundamentals, SQL or Python, a case study and behavioral questions. Timelines vary and can run long.
$135K
Avg. Base Comp
$153K
Avg. Total Comp
Not reported
Typical Rounds
3 months
Process Length
Candidates describe United Airlines Data Scientist interviews as a mix of resume and project discussion, machine learning and statistics fundamentals, and a case-style problem. The format is not identical across reports, so prepare for several shapes.
Project depth comes up repeatedly. One candidate had a 60-minute block on the statistics behind their projects, followed by a project presentation and a 30-minute case study in a single 2.5-hour in-person session. Another was asked to explain their thought process and the impact of what they had built. Be ready to defend your assumptions and statistical choices, not just your results.
On the technical side, bagging versus boosting appears in two reports, alongside Random Forest, RNNs and the metrics for classification and regression. One screen also included a Python string question and a three-table SQL join. A different candidate reported textbook-style questions, such as what an NP-hard problem is and convex versus concave functions. A candidate who received an offer for an operations research and optimization variant was asked about SQL basics, false positives versus false negatives, linear regression and explaining a technical concept to a non-technical listener.
One candidate received a case study 48 hours ahead with three options and chose one on allocating spare airplanes across airports. It was about reasoning rather than heavy technical work. Practice thinking aloud on operational problems.
Timing is the main frustration. One candidate reported a three-month process, with three to four weeks between stages. Another reported about three weeks to schedule each round. Plan for a slow loop, and follow up politely. Only one of the five reports ended in an offer.
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 United Airlines 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 United Airlines
A new airline came out as the fastest average boarding times compared to other airlines. What factors could have biased this result and what would you look into?
| Question | |
|---|---|
| Hurdles In Data Projects | |
| KNN From Scratch | |
| A Simpler KNN From Scratch | |
| Text Editor With OOP | |
| Employee Salaries | |
| Top 3 Users | |
| User Experience Percentage | |
| Distance Traveled | |
| Manager Team Sizes | |
| Download Facts | |
| Experiment Validity | |
| Weighted Keys | |
| Third Purchase | |
| Button AB Test | |
| Normal Distribution Sample | |
| Size of Joins | |
| Pre-Launching Shows | |
| Maximum Profit | |
| Bank Fraud Model | |
| Encoding Categorical Features | |
| WAU vs Open Rates | |
| Sum to N | |
| Network Experiment Design | |
| Assumptions of Linear Regression | |
| Revenue Retention | |
| Google Maps Improvement | |
| Type-ahead Search | |
| Demand Metrics | |
| Type I and II Errors |
Synthesized from candidate reports. Individual experiences may vary.
Candidates report starting with a recruiter or HR contact. One heard back within a week of applying, and another described an HR call to confirm scheduling and budget. These calls may be mostly logistical, though the reports do not describe them in detail, so keep your background summary and salary expectations ready.
Several candidates report that interviewers probed their resume projects. One described 60 minutes of resume and technical questions on the statistics behind their projects. Another was asked to explain their thought process and impact. Expect to defend your assumptions, methods and results, and in some loops to present a project.
Reported technical content includes bagging versus boosting, Random Forest, RNNs and classification and regression metrics. One candidate also had Python string manipulation and a three-table SQL join. Another reported theory questions on NP-hard problems and convex versus concave functions. Breadth appears to matter, so review fundamentals and be ready to explain trade-offs.
One candidate received a case study 48 hours in advance, with three options, and chose one about allocating spare airplanes across airports. Another had a 30-minute case study at the end of an in-person session. Candidates describe it as testing reasoning more than a perfect technical answer, so talk through assumptions and trade-offs.
One candidate was flown to Chicago headquarters for behavioral questions and team meetings. Another reported a 15-minute behavioral segment in an in-person session. Decision timing varied: one waited nearly a month, and another heard nothing for weeks. One report mentioned an offer, while the others ended in no offer or silence.
