One AstraZeneca data scientist candidate describes a recruiter screen, a HireVue video interview and a technical manager round on ML reasoning; another describes a half-day loop with ethics and a panel presentation.
$138K
Avg. Base Comp
$165K
Avg. Total Comp
Not reported
Typical Rounds
Not reported
Process Length
AstraZeneca's data scientist interviews, as candidates describe them, put machine learning reasoning and clear communication ahead of memorized definitions. A Supply Chain Data Scientist candidate started with a recruiter phone screen covering background, English proficiency and salary expectations, then recorded a HireVue video interview with up to three attempts per answer. The manager round that followed was technical: cross-validation, generative AI, what AI means to the candidate, and how they had adapted in past roles. The interviewer cared more about how the candidate explained their thinking than about perfect answers, and other team members asked how the candidate's background connected to the problems they work on. That candidate did not receive an offer.
A Principal Data Scientist candidate described a half-day loop that ended in an offer. The hiring manager and a Principal Data Scientist opened with machine learning fundamentals, including true and false positives in a clinical or biotech setting. A background discussion came next, then an ethics round, which the candidate connected to the company's work with drugs and patient data, and finally a presentation of data to a panel of around five scientists. The candidate noted that the exact presentation setup was unclear in their notes.
Two accounts are a small sample, and the team and seniority will shape the loop. Prepare to explain model choices and evaluation metrics in plain language, tie them to patient outcomes, and practice a short presentation for a mixed audience.
Synthesized from 7 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 Astrazeneca process.
The part that stood out most to me was how much of the process seemed to hinge on the early screening and the SHL-style assessment, because after that there wasn’t much clarity on next steps. I took the exam and felt pretty confident about it, especially since the returned result said I was likely to be a very good fit for the role compared to the comparison group, but I still never got a follow-up. I waited weeks and only found out later by checking the portal that I was no longer being considered, which was frustrating because I never received a rejection email.
Before that, the process itself was fairly standard on paper. There was a recruiter screen first, then a short phone call with the manager, and after that a longer interview block that sounded like a 2.5-hour session or, in another version of the process, a series of 30-minute rounds with the hiring manager and a small group panel. The questions were a mix of general experience and behavioral prompts, like what experience I had and how I handled pressure in previous work. On the technical side, they did ask about random forest modeling and tuning, and there was also a case-study style component with a presentation of the solution. The interviews came across as professional and friendly, and the turnaround for those who moved forward seemed pretty quick, but in my case the process ended without any real explanation. My main takeaway is to be ready for both the technical case presentation and the behavioral side, and to not assume the SHL result means you’re automatically moving on.
Prep tip from this candidate
Be ready to explain a random forest end-to-end, including how you would tune it, and practice presenting a case-study solution clearly. Also prepare for straightforward behavioral questions about handling pressure and describing your prior experience.
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 Astrazeneca
Write a function gcd to find the greatest common denominator of a list of integers
| Question | |
|---|---|
| Sum to Zero | |
| Hurdles In Data Projects | |
| Digit Accumulator | |
| Common Prefix | |
| Mapping Nicknames | |
| Client Solution Pushback | |
| Maximal Substring | |
| Regularization and Validation | |
| PCA and K-Means | |
| Automated Labeling | |
| 2nd Highest Salary | |
| Monthly Customer Report | |
| Cumulative Distribution | |
| Experiment Validity | |
| Last Transaction | |
| Weighted Keys | |
| Always Excited Users | |
| Brain Cancer Treatment Outcomes | |
| Retailer Data Warehouse | |
| Total Spent on Products | |
| P-value to a Layman | |
| RMS Error | |
| Reducing Error Margin | |
| Detecting ECG Tachycardia Runs | |
| Size of Joins | |
| Fair Coin | |
| Cumulative Reset | |
| Causal Email Journey | |
| Time Difference |
Synthesized from candidate reports. Individual experiences may vary.
The Supply Chain Data Scientist candidate's process opened with a phone screen. The recruiter walked through their background, assessed English proficiency and discussed salary expectations. The candidate called it standard, so have a concise summary of your experience and a salary expectation ready.
The same candidate then completed a pre-recorded HireVue interview. They got three attempts to record each response, which they found less intimidating than a live call because it left room to think through answers before committing. Practice structuring short spoken answers to common questions beforehand.
The manager round was described as genuinely technical. Questions covered cross-validation, generative AI, what AI means to the candidate and how they adapted to new situations. The interviewer valued how the candidate reasoned about ML concepts over perfect answers. Team members also asked how the candidate's background fit the team's problems.
The Principal Data Scientist candidate described a half-day loop. It started with ML fundamentals such as true and false positives in a clinical context, with the hiring manager and a Principal Data Scientist. It continued with a background discussion and an ethics round, and ended with a data presentation to a panel of about five scientists.
